Update compatibility target

Implement ListMultipartUploads, also refactor ListObjects and ListObjectsV2.

It took me some times as I wanted to propose the following things:
  - Using an iterator instead of the loop+goto pattern. I find it easier to read and it should enable some optimizations. For example, when consuming keys of a common prefix, we do many [redundant checks](https://git.deuxfleurs.fr/Deuxfleurs/garage/src/branch/main/src/api/s3_list.rs#L125-L156) while the only thing to do is to [check if the following key is still part of the common prefix](https://git.deuxfleurs.fr/Deuxfleurs/garage/src/branch/feature/s3-multipart-compat/src/api/s3_list.rs#L476).
  - Try to name things (see ExtractionResult and RangeBegin enums) and to separate concerns (see ListQuery and Accumulator)
  - An IO closure to make unit tests possibles.
  - Unit tests, to track regressions and document how to interact with the code
  - Integration tests with `s3api`. In the future, I would like to move them in Rust with the aws rust SDK.

Merging of the logic of ListMultipartUploads and ListObjects was not a goal but a consequence of the previous modifications.

Some points that we might want to discuss:
  - ListObjectsV1, when using pagination and delimiters, has a weird behavior (it lists multiple times the same prefix) with `aws s3api` due to the fact that it can not use our optimization to skip the whole prefix. It is independant from my refactor and can be tested with the commented `s3api` tests in `test-smoke.sh`. It probably has the same weird behavior on the official AWS S3 implementation.
  - Considering ListMultipartUploads, I had to "abuse" upload id marker to support prefix skipping. I send an `upload-id-marker` with the hardcoded value `include` to emulate your "including" token.
  - Some ways to test ListMultipartUploads with existing software (my tests are limited to s3api for now).

Co-authored-by: Quentin Dufour <quentin@deuxfleurs.fr>
Reviewed-on: https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/171
Co-authored-by: Quentin <quentin@dufour.io>
Co-committed-by: Quentin <quentin@dufour.io>

.drone.yml

@@ -6,103 +6,90 @@ workspace:
   base: /drone/garage
 
 volumes:
-- name: cargo_home
+- name: nix_store
+  host:
+    path: /var/lib/drone/nix
+- name: nix_config
   temp: {}
 
+environment:
+  HOME: /drone/garage
+
 steps:
-  - name: restore-cache
-    image: meltwater/drone-cache:dev
+  - name: setup nix
+    image: nixpkgs/nix:nixos-21.05
     volumes:
-    - name: cargo_home
-      path: /drone/cargo
-    environment:
-      AWS_ACCESS_KEY_ID:
-        from_secret: cache_aws_access_key_id
-      AWS_SECRET_ACCESS_KEY:
-        from_secret: cache_aws_secret_access_key
-    pull: true
-    settings:
-      restore: true
-      archive_format: "gzip"
-      bucket: drone-cache
-      cache_key: '{{ .Repo.Name }}_{{ checksum "Cargo.lock" }}_{{ arch }}_{{ os }}_gzip'
-      region: garage
-      mount:
-        - '/drone/cargo'
-        - 'target'
-      path_style: true
-      endpoint: https://garage.deuxfleurs.fr
-    when:
-      branch:
-      - nonexistent_skip_this_step
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - cp nix/nix.conf /etc/nix/nix.conf
+      - nix-build --no-build-output --no-out-link shell.nix --arg release false  -A inputDerivation
 
   - name: code quality
-    image: superboum/garage_builder_amd64:4
+    image: nixpkgs/nix:nixos-21.05
     volumes:
-    - name: cargo_home
-      path: /drone/cargo
-    environment:
-      CARGO_HOME: /drone/cargo
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
     commands:
-      - cargo fmt -- --check
-      - cargo clippy -- --deny warnings
+      - nix-shell --arg release false --run "cargo fmt -- --check"
+      - nix-shell --arg release false --run "cargo clippy -- --deny warnings"
 
   - name: build
-    image: superboum/garage_builder_amd64:4
+    image: nixpkgs/nix:nixos-21.05
     volumes:
-    - name: cargo_home
-      path: /drone/cargo
-    environment:
-      CARGO_HOME: /drone/cargo
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
     commands:
-      - pwd
-      - cargo build
+      - nix-build --no-build-output --argstr target x86_64-unknown-linux-musl --arg release false --argstr git_version $DRONE_COMMIT
 
-  - name: cargo-test
-    image: superboum/garage_builder_amd64:4
+  - name: unit tests
+    image: nixpkgs/nix:nixos-21.05
     volumes:
-    - name: cargo_home
-      path: /drone/cargo
-    environment:
-      CARGO_HOME: /drone/cargo
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
     commands:
-      - cargo test
-
-  - name: rebuild-cache
-    image: meltwater/drone-cache:dev
-    volumes:
-    - name: cargo_home
-      path: /drone/cargo
-    environment:
-      AWS_ACCESS_KEY_ID:
-        from_secret: cache_aws_access_key_id
-      AWS_SECRET_ACCESS_KEY:
-        from_secret: cache_aws_secret_access_key
-    pull: true
-    settings:
-      rebuild: true
-      archive_format: "gzip"
-      bucket: drone-cache
-      cache_key: '{{ .Repo.Name }}_{{ checksum "Cargo.lock" }}_{{ arch }}_{{ os }}_gzip'
-      region: garage
-      mount:
-        - '/drone/cargo'
-        - 'target'
-      path_style: true
-      endpoint: https://garage.deuxfleurs.fr
-    when:
-      branch:
-      - nonexistent_skip_this_step
+      - |
+        nix-build \
+          --no-build-output \
+          --argstr target x86_64-unknown-linux-musl \
+          --argstr compileMode test
+      - ./result*/bin/garage_api*
+      - ./result*/bin/garage_model*
+      - ./result*/bin/garage_rpc*
+      - ./result*/bin/garage_table*
+      - ./result*/bin/garage_util*
+      - ./result*/bin/garage_web*
+      - ./result*/bin/garage*
 
   - name: smoke-test
-    image: superboum/garage_builder_amd64:4
+    image: nixpkgs/nix:nixos-21.05
     volumes:
-    - name: cargo_home
-      path: /drone/cargo
-    environment:
-      CARGO_HOME: /drone/cargo
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
     commands:
-      - ./script/test-smoke.sh || (cat /tmp/garage.log; false)
+      - nix-build --no-build-output --argstr target x86_64-unknown-linux-musl --arg release false --argstr git_version $DRONE_COMMIT
+      - nix-shell --arg release false --run ./script/test-smoke.sh || (cat /tmp/garage.log; false)
+
+trigger:
+  event:
+  - custom
+  - push
+  - pull_request
+  - tag
+  - cron
+
+node:
+  nix: 1
 
 ---
 kind: pipeline
@@ -137,8 +124,393 @@ steps:
       repo:
         - Deuxfleurs/garage
 
+trigger:
+  event:
+  - custom
+  - push
+  - pull_request
+
+node:
+  nix: 1
+
+---
+kind: pipeline
+type: docker
+name: release-linux-x86_64
+
+volumes:
+- name: nix_store
+  host:
+    path: /var/lib/drone/nix
+- name: nix_config
+  temp: {}
+
+environment:
+  TARGET: x86_64-unknown-linux-musl
+
+steps:
+  - name: setup nix
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - cp nix/nix.conf /etc/nix/nix.conf
+      - nix-build --no-build-output --no-out-link shell.nix -A inputDerivation
+
+  - name: build
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - nix-build --no-build-output --argstr target $TARGET --arg release true --argstr git_version $DRONE_COMMIT
+
+  - name: integration
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - nix-shell --run ./script/test-smoke.sh || (cat /tmp/garage.log; false)
+
+  - name: push static binary
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    environment:
+      AWS_ACCESS_KEY_ID:
+        from_secret: garagehq_aws_access_key_id
+      AWS_SECRET_ACCESS_KEY:
+        from_secret: garagehq_aws_secret_access_key
+    commands:
+      - nix-shell --arg rust false --arg integration false --run "to_s3"
+
+  - name: docker build and publish
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    environment:
+      DOCKER_AUTH:
+        from_secret: docker_auth
+      DOCKER_PLATFORM: "linux/amd64"
+      CONTAINER_NAME: "dxflrs/amd64_garage"
+      HOME: "/kaniko"
+    commands:
+      - mkdir -p /kaniko/.docker
+      - echo $DOCKER_AUTH > /kaniko/.docker/config.json
+      - export CONTAINER_TAG=${DRONE_TAG:-$DRONE_COMMIT}
+      - nix-shell --arg rust false --arg integration false --run "to_docker"
+
+
+trigger:
+  event:
+  - promote
+  - cron
+
+node:
+  nix: 1
+
+---
+kind: pipeline
+type: docker
+name: release-linux-i686
+
+volumes:
+- name: nix_store
+  host:
+    path: /var/lib/drone/nix
+- name: nix_config
+  temp: {}
+
+environment:
+  TARGET: i686-unknown-linux-musl
+
+steps:
+  - name: setup nix
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - cp nix/nix.conf /etc/nix/nix.conf
+      - nix-build --no-build-output --no-out-link shell.nix -A inputDerivation
+
+  - name: build
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - nix-build --no-build-output --argstr target $TARGET --arg release true --argstr git_version $DRONE_COMMIT
+
+  - name: integration
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - nix-shell --run ./script/test-smoke.sh || (cat /tmp/garage.log; false)
+
+  - name: push static binary
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    environment:
+      AWS_ACCESS_KEY_ID:
+        from_secret: garagehq_aws_access_key_id
+      AWS_SECRET_ACCESS_KEY:
+        from_secret: garagehq_aws_secret_access_key
+    commands:
+      - nix-shell --arg rust false --arg integration false --run "to_s3"
+
+  - name: docker build and publish
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    environment:
+      DOCKER_AUTH:
+        from_secret: docker_auth
+      DOCKER_PLATFORM: "linux/386"
+      CONTAINER_NAME: "dxflrs/386_garage"
+      HOME: "/kaniko"
+    commands:
+      - mkdir -p /kaniko/.docker
+      - echo $DOCKER_AUTH > /kaniko/.docker/config.json
+      - export CONTAINER_TAG=${DRONE_TAG:-$DRONE_COMMIT}
+      - nix-shell --arg rust false --arg integration false --run "to_docker"
+
+trigger:
+  event:
+  - promote
+  - cron
+
+node:
+  nix: 1
+
+---
+kind: pipeline
+type: docker
+name: release-linux-aarch64
+
+volumes:
+- name: nix_store
+  host:
+    path: /var/lib/drone/nix
+- name: nix_config
+  temp: {}
+
+environment:
+  TARGET: aarch64-unknown-linux-musl
+
+steps:
+  - name: setup nix
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - cp nix/nix.conf /etc/nix/nix.conf
+      - nix-build --no-build-output --no-out-link ./shell.nix --arg rust false --arg integration false -A inputDerivation
+
+  - name: build
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - nix-build --no-build-output --argstr target $TARGET --arg release true --argstr git_version $DRONE_COMMIT
+
+  - name: push static binary
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    environment:
+      AWS_ACCESS_KEY_ID:
+        from_secret: garagehq_aws_access_key_id
+      AWS_SECRET_ACCESS_KEY:
+        from_secret: garagehq_aws_secret_access_key
+    commands:
+      - nix-shell --arg rust false --arg integration false --run "to_s3"
+
+  - name: docker build and publish
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    environment:
+      DOCKER_AUTH:
+        from_secret: docker_auth
+      DOCKER_PLATFORM: "linux/arm64"
+      CONTAINER_NAME: "dxflrs/arm64_garage"
+      HOME: "/kaniko"
+    commands:
+      - mkdir -p /kaniko/.docker
+      - echo $DOCKER_AUTH > /kaniko/.docker/config.json
+      - export CONTAINER_TAG=${DRONE_TAG:-$DRONE_COMMIT}
+      - nix-shell --arg rust false --arg integration false --run "to_docker"
+
+trigger:
+  event:
+  - promote
+  - cron
+
+node:
+  nix: 1
+
+---
+kind: pipeline
+type: docker
+name: release-linux-armv6l
+
+volumes:
+- name: nix_store
+  host:
+    path: /var/lib/drone/nix
+- name: nix_config
+  temp: {}
+
+environment:
+  TARGET: armv6l-unknown-linux-musleabihf
+
+steps:
+  - name: setup nix
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - cp nix/nix.conf /etc/nix/nix.conf
+      - nix-build --no-build-output --no-out-link --arg rust false --arg integration false -A inputDerivation
+
+  - name: build
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    commands:
+      - nix-build --no-build-output --argstr target $TARGET --arg release true --argstr git_version $DRONE_COMMIT
+
+  - name: push static binary
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    environment:
+      AWS_ACCESS_KEY_ID:
+        from_secret: garagehq_aws_access_key_id
+      AWS_SECRET_ACCESS_KEY:
+        from_secret: garagehq_aws_secret_access_key
+    commands:
+      - nix-shell --arg integration false --arg rust false --run "to_s3"
+
+  - name: docker build and publish
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    - name: nix_config
+      path: /etc/nix
+    environment:
+      DOCKER_AUTH:
+        from_secret: docker_auth
+      DOCKER_PLATFORM: "linux/arm"
+      CONTAINER_NAME: "dxflrs/arm_garage"
+      HOME: "/kaniko"
+    commands:
+      - mkdir -p /kaniko/.docker
+      - echo $DOCKER_AUTH > /kaniko/.docker/config.json
+      - export CONTAINER_TAG=${DRONE_TAG:-$DRONE_COMMIT}
+      - nix-shell --arg rust false --arg integration false --run "to_docker"
+
+trigger:
+  event:
+  - promote
+  - cron
+
+node:
+  nix: 1
+
+---
+kind: pipeline
+type: docker
+name: refresh-release-page
+
+volumes:
+- name: nix_store
+  host:
+    path: /var/lib/drone/nix
+
+steps:
+  - name: refresh-index
+    image: nixpkgs/nix:nixos-21.05
+    volumes:
+    - name: nix_store
+      path: /nix
+    environment:
+      AWS_ACCESS_KEY_ID:
+        from_secret: garagehq_aws_access_key_id
+      AWS_SECRET_ACCESS_KEY:
+        from_secret: garagehq_aws_secret_access_key
+    commands:
+      - mkdir -p /etc/nix && cp nix/nix.conf /etc/nix/nix.conf
+      - nix-shell --arg integration false --arg rust false --run "refresh_index"
+
+depends_on:
+  - release-linux-x86_64
+  - release-linux-i686
+  - release-linux-aarch64
+  - release-linux-armv6l
+
+trigger:
+  event:
+  - promote
+  - cron
+
+node:
+  nix: 1
+
 ---
 kind: signature
-hmac: e919f8a66d20ebfeeec56b291a8a0fdd59a482601da987fcf533d96d24768744
+hmac: 1c33490cc2902564c4250a409c156683d0d549b8c9d5aee4e46d1bde4e0ccf2c
 
 ...

Dockerfile

@@ -1,10 +1,7 @@
-FROM archlinux:latest
+FROM scratch
 
-RUN mkdir -p /garage/meta
-RUN mkdir -p /garage/data
 ENV RUST_BACKTRACE=1
 ENV RUST_LOG=garage=info
 
-COPY target/release/garage.stripped /garage/garage
-
-CMD /garage/garage server -c /garage/config.toml
+COPY result/bin/garage /
+CMD [ "/garage", "server"]

Makefile

@@ -1,5 +1,13 @@
+.PHONY: doc all release shell
+
 all:
 	clear; cargo build
 
+doc:
+	cd doc/book; mdbook build
+
 release:
-	RUSTFLAGS="-C link-arg=-fuse-ld=lld -C target-cpu=x86-64 -C target-feature=+sse2" cargo build --release --no-default-features
+	nix-build --arg release true
+
+shell:
+	nix-shell

README.md

@@ -3,10 +3,18 @@ Garage [![Build Status](https://drone.deuxfleurs.fr/api/badges/Deuxfleurs/garage
 
 <p align="center" style="text-align:center;">
 	<a href="https://garagehq.deuxfleurs.fr">
-	<img alt="Garage logo" src="doc/logo/garage.png" height="200" />
+	<img alt="Garage logo" src="https://garagehq.deuxfleurs.fr/img/logo.svg" height="200" />
 	</a>
 </p>
 
+<p align="center" style="text-align:center;">
+	[ <strong><a href="https://garagehq.deuxfleurs.fr/">Website and documentation</a></strong>
+	| <a href="https://garagehq.deuxfleurs.fr/_releases.html">Binary releases</a>
+	| <a href="https://git.deuxfleurs.fr/Deuxfleurs/garage">Git repository</a>
+	| <a href="https://matrix.to/#/%23garage:deuxfleurs.fr">Matrix channel</a>
+	]
+</p>
+
 Garage is a lightweight S3-compatible distributed object store, with the following goals:
 
 - As self-contained as possible
@@ -22,5 +30,3 @@ Non-goals include:
 - Erasure coding (our replication model is simply to copy the data as is on several nodes, in different datacenters if possible)
 
 Our main use case is to provide a distributed storage layer for small-scale self hosted services such as [Deuxfleurs](https://deuxfleurs.fr).
-
-**[Go to the documentation](https://garagehq.deuxfleurs.fr)**

TODO

@@ -1,27 +0,0 @@
-Testing
--------
-
-How are we going to test that our replication method works correctly?
-We will have to introduce lots of dummy data and then add/remove nodes many times.
-
-
-Attaining S3 compatibility
---------------------------
-
-- test multipart uploads
-- get ranges
-
-- fix sync not working in some cases ? (when starting from empty?)
-
-- api_server following the S3 semantics for head/get/put/list/delete: verify more that it works as intended
-- PUT requests: verify content-md5 if provided
-- possibly other necessary endpoints ?
-
-
-Lower priority
---------------
-
-- less a priority: hinted handoff
-- repair: re-propagate block ref table to rc
-- FIXME in rpc_server when garage shuts down and futures can be interrupted
-  (tokio::spawn should be replaced by a new function background::spawn_joinable)

default.nix

@@ -0,0 +1,86 @@
+{
+  system ? builtins.currentSystem,
+  release ? false,
+  target ? "x86_64-unknown-linux-musl",
+  compileMode ? null,
+  git_version ? null,
+}:
+
+with import ./nix/common.nix;
+
+let
+  crossSystem = { config = target; };
+in let
+  pkgs = import pkgsSrc {
+    inherit system crossSystem;
+    overlays = [ cargo2nixOverlay ];
+  };
+
+  /*
+   The following complexity should be abstracted by makePackageSet' (note the final quote).
+   However its code uses deprecated features of rust-overlay that can lead to bug.
+   Instead, we build our own rustChannel object with the recommended API of rust-overlay.
+   */
+  rustChannel = pkgs.rustPlatform.rust;
+
+  overrides = pkgs.buildPackages.rustBuilder.overrides.all ++ [
+    /*
+     We want to inject the git version while keeping the build deterministic.
+     As we do not want to consider the .git folder as part of the input source,
+     we ask the user (the CI often) to pass the value to Nix.
+    */
+    (pkgs.rustBuilder.rustLib.makeOverride {
+      name = "garage";
+      overrideAttrs = drv: if git_version != null then {
+        preConfigure = ''
+          ${drv.preConfigure or ""}
+          export GIT_VERSION="${git_version}"
+        '';
+      } else {};
+    })
+
+    /*
+     On a sandbox pure NixOS environment, /usr/bin/file is not available.
+     This is a known problem: https://github.com/NixOS/nixpkgs/issues/98440
+     We simply patch the file as suggested
+    */
+    /*(pkgs.rustBuilder.rustLib.makeOverride {
+      name = "libsodium-sys";
+      overrideAttrs = drv: {
+        preConfigure = ''
+          ${drv.preConfigure or ""}
+          sed -i 's,/usr/bin/file,${file}/bin/file,g' ./configure
+        '';
+      }
+    })*/
+  ];
+
+  packageFun = import ./Cargo.nix;
+
+  rustPkgs = pkgs.rustBuilder.makePackageSet {
+    inherit packageFun rustChannel release;
+    packageOverrides = overrides;
+
+    buildRustPackages = pkgs.buildPackages.rustBuilder.makePackageSet {
+      inherit rustChannel packageFun;
+      packageOverrides = overrides;
+    };
+
+    localPatterns = [
+      /*
+       The way the default rules are written make think we match recursively, on full path, but the rules are misleading.
+       In fact, the regex is only called on root elements of the crate (and not recursively).
+       This behavior does not work well with our nested modules.
+       We tried to build a "deny list" but negative lookup ahead are not supported on Nix.
+       As a workaround, we have to register all our submodules in this allow list...
+       */
+       ''^(src|tests)''          # fixed default
+       ''.*\.(rs|toml)$''        # fixed default
+       ''^(crdt|replication|cli|helper)''   # our crate submodules
+    ];
+  };
+
+in
+  if compileMode == "test"
+    then builtins.mapAttrs (name: value: rustPkgs.workspace.${name} { inherit compileMode; }) rustPkgs.workspace
+    else rustPkgs.workspace.garage { inherit compileMode; }

doc/book/src/SUMMARY.md

@@ -5,27 +5,45 @@
 - [Quick start](./quick_start/index.md)
 
 - [Cookbook](./cookbook/index.md)
-  - [Deploying Garage](./cookbook/real_world.md)
-  - [Configuring S3 clients](./cookbook/clients.md)
-  - [Hosting a website](./cookbook/website.md)
-  - [Recovering from failures](./cookbook/recovering.md)
+  - [Multi-node deployment](./cookbook/real_world.md)
   - [Building from source](./cookbook/from_source.md)
-  - [Starting with Systemd](./cookbook/systemd.md)
-  - [Integrate as a media backend]()
-  - [Operate a cluster]()
+  - [Integration with systemd](./cookbook/systemd.md)
+  - [Configuring a gateway node](./cookbook/gateways.md)
+  - [Exposing buckets as websites](./cookbook/exposing_websites.md)
+  - [Configuring a reverse proxy](./cookbook/reverse_proxy.md)
+  - [Recovering from failures](./cookbook/recovering.md)
+
+- [Integrations](./connect/index.md)
+  - [Browsing tools (awscli, mc...)](./connect/cli.md)
+  - [Apps (Nextcloud, Peertube...)](./connect/apps.md)
+  - [Websites (Hugo, Jekyll, Publii...)](./connect/websites.md)
+  - [Repositories (Docker, Nix, Git...)](./connect/repositories.md)
+  - [Backups (restic, duplicity...)](./connect/backup.md)
+  - [Your code (PHP, JS, Go...)](./connect/code.md)
+  - [FUSE (s3fs, goofys, s3backer...)](./connect/fs.md)
+
 
 - [Reference Manual](./reference_manual/index.md)
   - [Garage configuration file](./reference_manual/configuration.md)
+  - [Cluster layout management](./reference_manual/layout.md)
   - [Garage CLI](./reference_manual/cli.md)
   - [S3 compatibility status](./reference_manual/s3_compatibility.md)
 
 - [Design](./design/index.md)
-  - [Related Work](./design/related_work.md)
+  - [Goals and use Cases](./design/goals.md)
+  - [Benchmarks](./design/benchmarks.md)
+  - [Related work](./design/related_work.md)
   - [Internals](./design/internals.md)
 
 - [Development](./development/index.md)
   - [Setup your environment](./development/devenv.md)
-  - [Your first contribution]()
+  - [Development scripts](./development/scripts.md)
+  - [Release process](./development/release_process.md)
+  - [Miscellaneous notes](./development/miscellaneous_notes.md)
 
 - [Working Documents](./working_documents/index.md)
-  - [Load Balancing Data](./working_documents/load_balancing.md)
+  - [S3 compatibility target](./working_documents/compatibility_target.md)
+  - [Load balancing data](./working_documents/load_balancing.md)
+  - [Migrating from 0.5 to 0.6](./working_documents/migration_06.md)
+  - [Migrating from 0.3 to 0.4](./working_documents/migration_04.md)
+  - [Design draft](./working_documents/design_draft.md)

doc/book/src/connect/apps.md

@@ -0,0 +1,370 @@
+# Apps (Nextcloud, Peertube...)
+
+In this section, we cover the following web applications:
+
+| Name | Status | Note |
+|------|--------|------|
+| [Nextcloud](#nextcloud)     | ✅       |  Both Primary Storage and External Storage are supported    |
+| [Peertube](#peertube)     | ✅       | Must be configured with the website endpoint     |
+| [Mastodon](#mastodon)     | ❓       | Not yet tested     |
+| [Matrix](#matrix)     | ✅       |  Tested with `synapse-s3-storage-provider`    |
+| [Pixelfed](#pixelfed)     | ❓       |  Not yet tested    |
+| [Pleroma](#pleroma)     | ❓       |  Not yet tested    |
+| [Lemmy](#lemmy)     | ❓       |  Not yet tested    |
+| [Funkwhale](#funkwhale)     | ❓       | Not yet tested     |
+| [Misskey](#misskey)     | ❓       | Not yet tested     |
+| [Prismo](#prismo)     | ❓       | Not yet tested     |
+| [Owncloud OCIS](#owncloud-infinite-scale-ocis) |  ❓| Not yet tested |
+
+## Nextcloud
+
+Nextcloud is a popular file synchronisation and backup service.
+By default, Nextcloud stores its data on the local filesystem.
+If you want to expand your storage to aggregate multiple servers, Garage is the way to go.
+
+A S3 backend can be configured in two ways on Nextcloud, either as Primary Storage or as an External Storage.
+Primary storage will store all your data on S3, in an opaque manner, and will provide the best performances.
+External storage enable you to select which data will be stored on S3, your file hierarchy will be preserved in S3, but it might be slower.
+
+In the following, we cover both methods but before reading our guide, we suppose you have done some preliminary steps.
+First, we expect you have an already installed and configured Nextcloud instance.
+Second, we suppose you have created a key and a bucket.
+
+As a reminder, you can create a key for your nextcloud instance as follow:
+
+```bash
+garage key new --name nextcloud-key
+```
+
+Keep the Key ID and the Secret key in a pad, they will be needed later.  
+Then you can create a bucket and give read/write rights to your key on this bucket with:
+
+```bash
+garage bucket create nextcloud
+garage bucket allow nextcloud --read --write --key nextcloud-key
+```
+
+
+### Primary Storage
+
+Now edit your Nextcloud configuration file to enable object storage.
+On my installation, the config. file is located at the following path: `/var/www/nextcloud/config/config.php`.  
+We will add a new root key to the `$CONFIG` dictionnary named `objectstore`:
+
+```php
+<?php
+$CONFIG = array(
+/* your existing configuration */
+'objectstore' => [
+    'class' => '\\OC\\Files\\ObjectStore\\S3',
+    'arguments' => [
+        'bucket' => 'nextcloud',   // Your bucket name, must be created before
+        'autocreate' => false,     // Garage does not support autocreate
+        'key'    => 'xxxxxxxxx',   // The Key ID generated previously
+        'secret' => 'xxxxxxxxx',   // The Secret key generated previously
+        'hostname' => '127.0.0.1', // Can also be a domain name, eg. garage.example.com
+        'port' => 3900,            // Put your reverse proxy port or your S3 API port
+        'use_ssl' => false,        // Set it to true if you have a TLS enabled reverse proxy
+        'region' => 'garage',      // Garage has only one region named "garage"
+        'use_path_style' => true   // Garage supports only path style, must be set to true
+    ],
+],
+```
+
+That's all, your Nextcloud will store all your data to S3.
+To test your new configuration, just reload your Nextcloud webpage and start sending data.
+
+*External link:* [Nextcloud Documentation > Primary Storage](https://docs.nextcloud.com/server/latest/admin_manual/configuration_files/primary_storage.html)
+
+### External Storage
+
+**From the GUI.** Activate the "External storage support" app from the "Applications" page (click on your account icon on the top right corner of your screen to display the menu). Go to your parameters page (also located below your account icon). Click on external storage (or the corresponding translation in your language).
+
+[![Screenshot of the External Storage form](./cli-nextcloud-gui.png)](./cli-nextcloud-gui.png)
+*Click on the picture to zoom*
+
+Add a new external storage. Put what you want in "folder name" (eg. "shared"). Select "Amazon S3". Keep "Access Key" for the Authentication field.
+In Configuration, put your bucket name (eg. nextcloud), the host (eg. 127.0.0.1), the port (eg. 3900 or 443), the region (garage). Tick the SSL box if you have put an HTTPS proxy in front of garage. You must tick the "Path access" box and you must leave the "Legacy authentication (v2)" box empty. Put your Key ID (eg. GK...) and your Secret Key in the last two input boxes. Finally click on the tick symbol on the right of your screen.
+
+Now go to your "Files" app and a new "linked folder" has appeared with the name you chose earlier (eg. "shared").
+
+*External link:* [Nextcloud Documentation > External Storage Configuration GUI](https://docs.nextcloud.com/server/latest/admin_manual/configuration_files/external_storage_configuration_gui.html)
+
+**From the CLI.** First install the external storage application:
+
+```bash
+php occ app:install files_external
+```
+
+Then add a new mount point with:
+
+```bash
+ php occ files_external:create \
+  -c bucket=nextcloud \
+  -c hostname=127.0.0.1 \
+  -c port=3900 \
+  -c region=garage \
+  -c use_ssl=false \
+  -c use_path_style=true \
+  -c legacy_auth=false \
+  -c key=GKxxxx \
+  -c secret=xxxx \
+  shared amazons3 amazons3::accesskey
+```
+
+Adapt the `hostname`, `port`, `use_ssl`, `key`, and `secret` entries to your configuration.
+Do not change the `use_path_style` and `legacy_auth` entries, other configurations are not supported.
+
+*External link:* [Nextcloud Documentation > occ command > files external](https://docs.nextcloud.com/server/latest/admin_manual/configuration_server/occ_command.html#files-external-label)
+
+
+## Peertube
+
+Peertube proposes a clever integration of S3 by directly exposing its endpoint instead of proxifying requests through the application.
+In other words, Peertube is only responsible of the "control plane" and offload the "data plane" to Garage.
+In return, this system is a bit harder to configure.
+We show how it is still possible to configure Garage with Peertube, allowing you to spread the load and the bandwidth usage on the Garage cluster.
+
+
+### Create resources in Garage
+
+Create a key for Peertube:
+
+```bash
+garage key new --name peertube-key
+```
+
+Keep the Key ID and the Secret key in a pad, they will be needed later.  
+
+We need two buckets, one for normal videos (named peertube-video) and one for webtorrent videos (named peertube-playlist).
+```bash
+garage bucket create peertube-video
+garage bucket create peertube-playlist
+```
+
+Now we allow our key to read and write on these buckets:
+
+```
+garage bucket allow peertube-playlists --read --write --owner --key peertube-key
+garage bucket allow peertube-videos --read --write --owner --key peertube-key
+```
+
+We also need to expose these buckets publicly to serve their content to users:
+
+```bash
+garage bucket website --allow peertube-playlists
+garage bucket website --allow peertube-videos
+```
+
+Finally, we must allow Cross-Origin Resource Sharing (CORS). 
+CORS are required by your browser to allow requests triggered from the peertube website (eg. peertube.tld) to your bucket's domain (eg. peertube-videos.web.garage.tld)
+
+```bash
+export CORS='{"CORSRules":[{"AllowedHeaders":["*"],"AllowedMethods":["GET"],"AllowedOrigins":["*"]}]}'
+aws --endpoint http://s3.garage.localhost s3api put-bucket-cors --bucket peertube-playlists --cors-configuration $CORS
+aws --endpoint http://s3.garage.localhost s3api put-bucket-cors --bucket peertube-videos --cors-configuration $CORS
+```
+
+These buckets are now accessible on the web port (by default 3902) with the following URL: `http://<bucket><root_domain>:<web_port>` where the root domain is defined in your configuration file (by default `.web.garage`). So we have currently the following URLs: 
+  * http://peertube-playlists.web.garage:3902
+  * http://peertube-videos.web.garage:3902
+
+Make sure you (will) have a corresponding DNS entry for them.
+
+### Configure Peertube
+
+You must edit the file named `config/production.yaml`, we are only modifying the root key named `object_storage`:
+
+```yaml
+object_storage:
+  enabled: true
+
+  # Put localhost only if you have a garage instance running on that node
+  endpoint: 'http://localhost:3900' # or "garage.example.com" if you have TLS on port 443
+
+  # Garage supports only one region for now, named garage
+  region: 'garage'
+
+  credentials:
+    access_key_id: 'GKxxxx'
+    secret_access_key: 'xxxx'
+
+  max_upload_part: 2GB
+
+  streaming_playlists:
+    bucket_name: 'peertube-playlist'
+
+    # Keep it empty for our example
+    prefix: ''
+
+    # You must fill this field to make Peertube use our reverse proxy/website logic
+    base_url: 'http://peertube-playlists.web.garage.localhost' # Example: 'https://mirror.example.com'
+
+  # Same settings but for webtorrent videos
+  videos:
+    bucket_name: 'peertube-video'
+    prefix: ''
+    # You must fill this field to make Peertube use our reverse proxy/website logic
+    base_url: 'http://peertube-videos.web.garage.localhost'
+```
+
+### That's all
+
+Everything must be configured now, simply restart Peertube and try to upload a video.
+
+Peertube will start by serving the video from its own domain while it is encoding.
+Once the encoding is done, the video is uploaded to Garage.
+You can now reload the page and see in your browser console that data are fetched directly from your bucket.
+
+*External link:* [Peertube Documentation > Remote Storage](https://docs.joinpeertube.org/admin-remote-storage)
+
+## Mastodon
+
+https://docs.joinmastodon.org/admin/config/#cdn
+
+## Matrix
+
+Matrix is a chat communication protocol. Its main stable server implementation, [Synapse](https://matrix-org.github.io/synapse/latest/), provides a module to store media on a S3 backend. Additionally, a server independent media store supporting S3 has been developped by the community, it has been made possible thanks to how the matrix API has been designed and will work with implementations like Conduit, Dendrite, etc.
+
+### synapse-s3-storage-provider (synapse only)
+
+Supposing you have a working synapse installation, you can add the module with pip:
+
+```bash
+ pip3 install --user git+https://github.com/matrix-org/synapse-s3-storage-provider.git
+```
+
+Now create a bucket and a key for your matrix instance (note your Key ID and Secret Key somewhere, they will be needed later):
+
+```bash
+garage key new --name matrix-key
+garage bucket create matrix
+garage bucket allow matrix --read --write --key matrix-key
+```
+
+Then you must edit your server configuration (eg. `/etc/matrix-synapse/homeserver.yaml`) and add the `media_storage_providers` root key:
+
+```yaml
+media_storage_providers:
+- module: s3_storage_provider.S3StorageProviderBackend
+  store_local: True    # do we want to store on S3 media created by our users?
+  store_remote: True   # do we want to store on S3 media created
+                       # by users of others servers federated to ours?
+  store_synchronous: True  # do we want to wait that the file has been written before returning?
+  config:
+    bucket: matrix       # the name of our bucket, we chose matrix earlier
+    region_name: garage  # only "garage" is supported for the region field
+    endpoint_url: http://localhost:3900 # the path to the S3 endpoint
+    access_key_id: "GKxxx" # your Key ID
+    secret_access_key: "xxxx" # your Secret Key
+```
+
+Note that uploaded media will also be stored locally and this behavior can not be deactivated, it is even required for
+some operations like resizing images.
+In fact, your local filesysem is considered as a cache but without any automated way to garbage collect it.
+
+We can build our garbage collector with `s3_media_upload`, a tool provided with the module.
+If you installed the module with the command provided before, you should be able to bring it in your path:
+
+```
+PATH=$HOME/.local/bin/:$PATH
+command -v s3_media_upload
+```
+
+Now we can write a simple script (eg `~/.local/bin/matrix-cache-gc`):
+
+```bash
+#!/bin/bash
+
+## CONFIGURATION ##
+AWS_ACCESS_KEY_ID=GKxxx
+AWS_SECRET_ACCESS_KEY=xxxx
+S3_ENDPOINT=http://localhost:3900
+S3_BUCKET=matrix
+MEDIA_STORE=/var/lib/matrix-synapse/media
+PG_USER=matrix
+PG_PASS=xxxx
+PG_DB=synapse
+PG_HOST=localhost
+PG_PORT=5432
+
+## CODE ##
+PATH=$HOME/.local/bin/:$PATH
+cat > database.yaml <<EOF 
+user: $PG_USER
+password: $PG_PASS
+database: $PG_DB
+host: $PG_HOST
+port: $PG_PORT
+EOF
+
+s3_media_upload update-db 1d
+s3_media_upload --no-progress check-deleted $MEDIA_STORE
+s3_media_upload --no-progress upload $MEDIA_STORE $S3_BUCKET --delete --endpoint-url $S3_ENDPOINT
+```
+
+This script will list all the medias that were not accessed in the 24 hours according to your database.
+It will check if, in this list, the file still exists in the local media store.
+For files that are still in the cache, it will upload them to S3 if they are not already present (in case of a crash or an initial synchronisation).
+Finally, the script will delete these files from the cache.
+
+Make this script executable and check that it works:
+
+```bash
+chmod +x $HOME/.local/bin/matrix-cache-gc
+matrix-cache-gc
+```
+
+Add it to your crontab. Open the editor with:
+
+```bash
+crontab -e
+```
+
+And add a new line. For example, to run it every 10 minutes:
+
+```cron
+*/10 * * * * $HOME/.local/bin/matrix-cache-gc
+```
+
+*External link:* [Github > matrix-org/synapse-s3-storage-provider](https://github.com/matrix-org/synapse-s3-storage-provider)
+
+### matrix-media-repo (server independent)
+
+*External link:* [matrix-media-repo Documentation > S3](https://docs.t2bot.io/matrix-media-repo/configuration/s3-datastore.html)
+
+## Pixelfed
+
+[Pixelfed Technical Documentation > Configuration](https://docs.pixelfed.org/technical-documentation/env.html#filesystem)
+
+## Pleroma
+
+[Pleroma Documentation > Pleroma.Uploaders.S3](https://docs-develop.pleroma.social/backend/configuration/cheatsheet/#pleromauploaderss3)
+
+## Lemmy
+
+Lemmy uses pict-rs that [supports S3 backends](https://git.asonix.dog/asonix/pict-rs/commit/f9f4fc63d670f357c93f24147c2ee3e1278e2d97)
+
+## Funkwhale
+
+[Funkwhale Documentation > S3 Storage](https://docs.funkwhale.audio/admin/configuration.html#s3-storage)
+
+## Misskey
+
+[Misskey Github > commit 9d94424](https://github.com/misskey-dev/misskey/commit/9d944243a3a59e8880a360cbfe30fd5a3ec8d52d)
+
+## Prismo
+
+[Prismo Gitlab > .env.production.sample](https://gitlab.com/prismosuite/prismo/-/blob/dev/.env.production.sample#L26-33)
+
+## Owncloud Infinite Scale (ocis)
+
+OCIS could be compatible with S3:
+  - [Deploying OCIS with S3](https://owncloud.dev/ocis/deployment/ocis_s3/)
+  - [OCIS 1.7 release note](https://central.owncloud.org/t/owncloud-infinite-scale-tech-preview-1-7-enables-s3-storage/32514/3)
+
+## Unsupported
+
+  - Mobilizon: No S3 integration
+  - WriteFreely: No S3 integration
+  - Plume: No S3 integration

doc/book/src/connect/backup.md

@@ -0,0 +1,33 @@
+# Backups (restic, duplicity...)
+
+Backups are essential for disaster recovery but they are not trivial to manage.
+Using Garage as your backup target will enable you to scale your storage as needed while ensuring high availability.
+
+## Borg Backup
+
+Borg Backup is very popular among the backup tools but it is not yet compatible with the S3 API.
+We recommend using any other tool listed in this guide because they are all compatible with the S3 API.
+If you still want to use Borg, you can use it with `rclone mount`.
+
+
+
+## Restic
+
+*External links:* [Restic Documentation > Amazon S3](https://restic.readthedocs.io/en/stable/030_preparing_a_new_repo.html#amazon-s3)
+
+## Duplicity
+
+*External links:* [Duplicity > man](https://duplicity.gitlab.io/duplicity-web/vers8/duplicity.1.html) (scroll to "URL Format" and "A note on Amazon S3")
+
+## Duplicati
+
+*External links:* [Duplicati Documentation > Storage Providers](https://github.com/kees-z/DuplicatiDocs/blob/master/docs/05-storage-providers.md#s3-compatible)
+
+## knoxite
+
+*External links:* [Knoxite Documentation > Storage Backends](https://knoxite.com/docs/storage-backends/#amazon-s3)
+
+## kopia
+
+*External links:* [Kopia Documentation > Repositories](https://kopia.io/docs/repositories/#amazon-s3)
+

doc/book/src/connect/cli.md

@@ -0,0 +1,277 @@
+# Browsing tools
+
+Browsing tools allow you to query the S3 API without too many abstractions.
+These tools are particularly suitable for debug, backups, website deployments or any scripted task that need to handle data.
+
+| Name | Status | Note |
+|------|--------|------|
+| [Minio client](#minio-client-recommended)     | ✅       |  Recommended  |
+| [AWS CLI](#aws-cli)     | ✅       | Recommended   |
+| [rclone](#rclone)     | ✅       |    |
+| [s3cmd](#s3cmd)     | ✅       |    |
+| [(Cyber)duck](#cyberduck--duck)     | ✅       |    |
+| [WinSCP (libs3)](#winscp)     | ✅       | No instructions yet   |
+
+
+## Minio client
+
+Use the following command to set an "alias", i.e. define a new S3 server to be
+used by the Minio client:
+
+```bash
+mc alias set \
+  garage \
+  <endpoint> \
+  <access key> \
+  <secret key> \
+  --api S3v4
+```
+
+Remember that `mc` is sometimes called `mcli` (such as on Arch Linux), to avoid conflicts
+with Midnight Commander.
+
+Some commands:
+
+```bash
+# list buckets
+mc ls garage/
+
+# list objets in a bucket
+mc ls garage/my_files
+
+# copy from your filesystem to garage
+mc cp /proc/cpuinfo garage/my_files/cpuinfo.txt
+
+# copy from garage to your filesystem
+mc cp garage/my_files/cpuinfo.txt /tmp/cpuinfo.txt
+
+# mirror a folder from your filesystem to garage
+mc mirror --overwrite ./book garage/garagehq.deuxfleurs.fr
+```
+
+
+## AWS CLI
+
+Create a file named `~/.aws/credentials` and put:
+
+```toml
+[default]
+aws_access_key_id=xxxx
+aws_secret_access_key=xxxx
+```
+
+Then a file named `~/.aws/config` and put:
+
+```toml
+[default]
+region=garage
+```
+
+Now, supposing Garage is listening on `http://127.0.0.1:3900`, you can list your buckets with:
+
+```bash
+aws --endpoint-url http://127.0.0.1:3900 s3 ls
+```
+
+Passing the `--endpoint-url` parameter to each command is annoying but AWS developers do not provide a corresponding configuration entry.
+As a workaround, you can redefine the aws command by editing the file `~/.bashrc`:
+
+```
+function aws { command aws --endpoint-url http://127.0.0.1:3900 $@ ; }
+```
+
+*Do not forget to run `source ~/.bashrc` or to start a new terminal before running the next commands.*
+
+Now you can simply run:
+
+```bash
+# list buckets
+aws s3 ls
+
+# list objects of a bucket
+aws s3 ls s3://my_files
+
+# copy from your filesystem to garage
+aws s3 cp /proc/cpuinfo s3://my_files/cpuinfo.txt
+
+# copy from garage to your filesystem
+aws s3 cp s3/my_files/cpuinfo.txt /tmp/cpuinfo.txt
+```
+
+## `rclone`
+
+`rclone` can be configured using the interactive assistant invoked using `rclone config`.
+
+You can also configure `rclone` by writing directly its configuration file.
+Here is a template `rclone.ini` configuration file (mine is located at `~/.config/rclone/rclone.conf`):
+
+```ini
+[garage]
+type = s3
+provider = Other
+env_auth = false
+access_key_id = <access key>
+secret_access_key = <secret key>
+region = <region>
+endpoint = <endpoint>
+force_path_style = true
+acl = private
+bucket_acl = private
+```
+
+Now you can run:
+
+```bash
+# list buckets
+rclone lsd garage:
+
+# list objects of a bucket aggregated in directories
+rclone lsd garage:my-bucket
+
+# copy from your filesystem to garage
+echo hello world > /tmp/hello.txt
+rclone copy /tmp/hello.txt garage:my-bucket/
+
+# copy from garage to your filesystem
+rclone copy garage:quentin.divers/hello.txt .
+
+# see all available subcommands
+rclone help
+```
+
+## `s3cmd`
+
+Here is a template for the `s3cmd.cfg` file to talk with Garage:
+
+```ini
+[default]
+access_key = <access key>
+secret_key = <secret key>
+host_base = <endpoint without http(s)://>
+host_bucket = <same as host_base>
+use_https = <False or True>
+```
+
+And use it as follow:
+
+```bash
+# List buckets
+s3cmd ls
+
+# s3cmd objects inside a bucket
+s3cmd ls s3://my-bucket
+
+# copy from your filesystem to garage
+echo hello world > /tmp/hello.txt
+s3cmd put /tmp/hello.txt s3://my-bucket/
+
+# copy from garage to your filesystem
+s3cmd get s3://my-bucket/hello.txt hello.txt
+```
+
+## Cyberduck & duck
+
+Both Cyberduck (the GUI) and duck (the CLI) have a concept of "Connection Profiles" that contain some presets for a specific provider.
+We wrote the following connection profile for Garage:
+
+```xml
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
+<plist version="1.0">
+    <dict>
+        <key>Protocol</key>
+        <string>s3</string>
+        <key>Vendor</key>
+        <string>garage</string>
+        <key>Scheme</key>
+        <string>https</string>
+        <key>Description</key>
+        <string>GarageS3</string>
+        <key>Default Hostname</key>
+        <string>127.0.0.1</string>
+        <key>Default Port</key>
+        <string>4443</string>
+        <key>Hostname Configurable</key>
+        <false/>
+        <key>Port Configurable</key>
+        <false/>
+        <key>Username Configurable</key>
+        <true/>
+        <key>Username Placeholder</key>
+        <string>Access Key ID (GK...)</string>
+        <key>Password Placeholder</key>
+        <string>Secret Key</string>
+        <key>Properties</key>
+        <array>
+            <string>s3service.disable-dns-buckets=true</string>
+        </array>
+        <key>Region</key>
+        <string>garage</string>
+        <key>Regions</key>
+        <array>
+            <string>garage</string>
+        </array>
+    </dict>
+</plist>
+```
+
+*Note: If your garage instance is configured with vhost access style, you can remove `s3service.disable-dns-buckets=true`.*
+
+### Instructions for the GUI
+
+Copy the connection profile, and save it anywhere as `garage.cyberduckprofile`.
+Then find this file with your file explorer and double click on it: Cyberduck will open a connection wizard for this profile.
+Simply follow the wizard and you should be done!
+
+### Instuctions for the CLI
+
+To configure duck (Cyberduck's CLI tool), start by creating its folder hierarchy:
+
+```
+mkdir -p ~/.duck/profiles/
+```
+
+Then, save the connection profile for Garage in `~/.duck/profiles/garage.cyberduckprofile`.
+To set your credentials in `~/.duck/credentials`, use the following commands to generate the appropriate string:
+
+```bash
+export AWS_ACCESS_KEY_ID="GK..."
+export AWS_SECRET_ACCESS_KEY="..."
+export HOST="s3.garage.localhost"
+export PORT="4443"
+export PROTOCOL="https"
+
+cat > ~/.duck/credentials <<EOF
+$PROTOCOL\://$AWS_ACCESS_KEY_ID@$HOST\:$PORT=$AWS_SECRET_ACCESS_KEY
+EOF
+```
+
+And finally, I recommend appending a small wrapper to your `~/.bashrc` to avoid setting the username on each command (do not forget to replace `GK...` by your access key):
+
+```bash
+function duck { command duck --username GK... $@ ; }
+```
+
+Finally, you can then use `duck` as follow:
+
+```bash
+# List buckets
+duck --list garage:/
+
+# List objects in a bucket
+duck --list garage:/my-files/
+
+# Download an object
+duck --download garage:/my-files/an-object.txt /tmp/object.txt
+
+# Upload an object
+duck --upload /tmp/object.txt garage:/my-files/another-object.txt
+
+# Delete an object
+duck --delete garage:/my-files/an-object.txt
+```
+
+## WinSCP (libs3)
+
+*No instruction yet. You can find ones in french [in our wiki](https://wiki.deuxfleurs.fr/fr/Guide/Garage/WinSCP).*
+

doc/book/src/connect/code.md

@@ -0,0 +1,79 @@
+# Your code (PHP, JS, Go...)
+
+If you are developping a new application, you may want to use Garage to store your user's media.
+
+The S3 API that Garage uses is a standard REST API, so as long as you can make HTTP requests,
+you can query it. You can check the [S3 REST API Reference](https://docs.aws.amazon.com/AmazonS3/latest/API/API_Operations_Amazon_Simple_Storage_Service.html) from Amazon to learn more.
+
+Developping your own wrapper around the REST API is time consuming and complicated.
+Instead, there are some libraries already avalaible.
+
+Some of them are maintained by Amazon, some by Minio, others by the community.
+
+## PHP
+
+  - Amazon aws-sdk-php
+    - [Installation](https://docs.aws.amazon.com/sdk-for-php/v3/developer-guide/getting-started_installation.html)
+    - [Reference](https://docs.aws.amazon.com/aws-sdk-php/v3/api/api-s3-2006-03-01.html)
+    - [Example](https://docs.aws.amazon.com/sdk-for-php/v3/developer-guide/s3-examples-creating-buckets.html)
+
+## Javascript
+
+  - Minio SDK
+    - [Reference](https://docs.min.io/docs/javascript-client-api-reference.html)
+
+  - Amazon aws-sdk-js
+    - [Installation](https://docs.aws.amazon.com/sdk-for-javascript/v3/developer-guide/getting-started.html)
+    - [Reference](https://docs.aws.amazon.com/AWSJavaScriptSDK/latest/AWS/S3.html)
+    - [Example](https://docs.aws.amazon.com/sdk-for-javascript/v3/developer-guide/s3-example-creating-buckets.html)
+
+## Golang
+
+  - Minio minio-go-sdk
+    - [Reference](https://docs.min.io/docs/golang-client-api-reference.html)
+
+  - Amazon aws-sdk-go-v2
+    - [Installation](https://aws.github.io/aws-sdk-go-v2/docs/getting-started/)
+    - [Reference](https://pkg.go.dev/github.com/aws/aws-sdk-go-v2/service/s3)
+    - [Example](https://aws.github.io/aws-sdk-go-v2/docs/code-examples/s3/putobject/)
+
+## Python
+
+  - Minio SDK
+    - [Reference](https://docs.min.io/docs/python-client-api-reference.html)
+
+  - Amazon boto3
+    - [Installation](https://boto3.amazonaws.com/v1/documentation/api/latest/guide/quickstart.html)
+    - [Reference](https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/s3.html)
+    - [Example](https://boto3.amazonaws.com/v1/documentation/api/latest/guide/s3-uploading-files.html)
+
+## Java
+
+  - Minio SDK
+    - [Reference](https://docs.min.io/docs/java-client-api-reference.html)
+
+  - Amazon aws-sdk-java
+    - [Installation](https://docs.aws.amazon.com/sdk-for-java/latest/developer-guide/get-started.html)
+    - [Reference](https://sdk.amazonaws.com/java/api/latest/software/amazon/awssdk/services/s3/S3Client.html)
+    - [Example](https://docs.aws.amazon.com/sdk-for-java/latest/developer-guide/examples-s3-objects.html)
+
+## Rust
+
+  - Amazon aws-rust-sdk
+    - [Github](https://github.com/awslabs/aws-sdk-rust)
+
+## .NET
+
+  - Minio SDK
+    - [Reference](https://docs.min.io/docs/dotnet-client-api-reference.html)
+
+  - Amazon aws-dotnet-sdk
+
+## C++
+
+  - Amazon aws-cpp-sdk
+
+## Haskell
+
+  - Minio SDK
+    - [Reference](https://docs.min.io/docs/haskell-client-api-reference.html)

doc/book/src/connect/fs.md

@@ -0,0 +1,68 @@
+# FUSE (s3fs, goofys, s3backer...)
+
+**WARNING! Garage is not POSIX compatible.
+Mounting S3 buckets as filesystems will not provide POSIX compatibility.
+If you are not careful, you will lose or corrupt your data.**
+
+Do not use these FUSE filesystems to store any database files (eg. MySQL, Postgresql, Mongo or sqlite),
+any daemon cache (dovecot, openldap, gitea, etc.),
+and more generally any software that use locking, advanced filesystems features or make any synchronisation assumption.
+Ideally, avoid these solutions at all for any serious or production use.
+
+## rclone mount
+
+rclone uses the same configuration when used [in CLI](/connect/cli.html) and mount mode.
+We suppose you have the following entry in your `rclone.ini` (mine is located in `~/.config/rclone/rclone.conf`):
+
+```toml
+[garage]
+type = s3
+provider = Other
+env_auth = false
+access_key_id = <access key>
+secret_access_key = <secret key>
+region = <region>
+endpoint = <endpoint>
+force_path_style = true
+acl = private
+bucket_acl = private
+```
+
+Then you can mount and access any bucket as follow:
+
+```bash
+# mount the bucket
+mkdir /tmp/my-bucket
+rclone mount --daemon garage:my-bucket /tmp/my-bucket
+
+# set your working directory to the bucket
+cd /tmp/my-bucket
+
+# create a file
+echo hello world > hello.txt
+
+# access the file
+cat hello.txt
+
+# unmount the bucket
+cd
+fusermount -u /tmp/my-bucket
+```
+
+*External link:* [rclone documentation > rclone mount](https://rclone.org/commands/rclone_mount/)
+
+## s3fs
+
+*External link:* [s3fs github > README.md](https://github.com/s3fs-fuse/s3fs-fuse#examples)
+
+## goofys
+
+*External link:* [goofys github > README.md](https://github.com/kahing/goofys#usage)
+
+## s3backer
+
+*External link:* [s3backer github > manpage](https://github.com/archiecobbs/s3backer/wiki/ManPage)
+
+## csi-s3
+
+*External link:* [csi-s3 Github > README.md](https://github.com/ctrox/csi-s3)

doc/book/src/connect/index.md

@@ -0,0 +1,42 @@
+# Integrations
+
+Garage implements the Amazon S3 protocol, which makes it compatible with many existing software programs.
+
+In particular, you will find here instructions to connect it with:
+
+  - [Browsing tools](./cli.md)
+  - [Applications](./apps.md)
+  - [Website hosting](./websites.md)
+  - [Software repositories](./repositories.md)
+  - [Your own code](./code.md)
+  - [FUSE](./fs.md)
+
+### Generic instructions
+
+To configure S3-compatible software to interact with Garage,
+you will need the following parameters:
+
+- An **API endpoint**: this corresponds to the HTTP or HTTPS address
+  used to contact the Garage server. When runing Garage locally this will usually
+  be `http://127.0.0.1:3900`. In a real-world setting, you would usually have a reverse-proxy
+  that adds TLS support and makes your Garage server available under a public hostname
+  such as `https://garage.example.com`.
+
+- An **API access key** and its associated **secret key**. These usually look something
+  like this: `GK3515373e4c851ebaad366558` (access key),
+  `7d37d093435a41f2aab8f13c19ba067d9776c90215f56614adad6ece597dbb34` (secret key).
+  These keys are created and managed using the `garage` CLI, as explained in the
+  [quick start](../quick_start/index.md) guide.
+
+Most S3 clients can be configured easily with these parameters,
+provided that you follow the following guidelines:
+
+- **Be careful to DNS-style/path-style access:** Garage supports both DNS-style buckets, which are now by default
+  on Amazon S3, and legacy path-style buckets. If you use a reverse proxy in front of Garage, 
+  make sure that you configured it to support the access-style required by the software you want to use.
+
+- **Configuring the S3 region:** Garage requires your client to talk to the correct "S3 region",
+  which is set in the configuration file. This is often set just to `garage`.
+  If this is not configured explicitly, clients usually try to talk to region `us-east-1`.
+  Garage should normally redirect your client to the correct region,
+  but in case your client does not support this you might have to configure it manually.

doc/book/src/connect/repositories.md

@@ -0,0 +1,208 @@
+# Repositories (Docker, Nix, Git...)
+
+Whether you need to store and serve binary packages or source code, you may want to deploy a tool referred as a repository or registry.
+Garage can also help you serve this content.
+
+| Name | Status | Note |
+|------|--------|------|
+| [Gitea](#gitea)     | ✅       |   |
+| [Docker](#generic-static-site-generator)     | ✅        | Requires garage >= v0.6.0   |
+| [Nix](#generic-static-site-generator)     | ✅        |     |
+| [Gitlab](#gitlab)     |  ❓        |  Not yet tested    |
+
+
+
+## Gitea
+
+You can use Garage with Gitea to store your [git LFS](https://git-lfs.github.com/) data, your users' avatar, and their attachements.
+You can configure a different target for each data type (check `[lfs]` and `[attachment]` sections of the Gitea documentation) and you can provide a default one through the `[storage]` section.
+
+Let's start by creating a key and a bucket (your key id and secret will be needed later, keep them somewhere):
+
+```bash
+garage key new --name gitea-key
+garage bucket create gitea
+garage bucket allow gitea --read --write --key gitea-key
+```
+
+Then you can edit your configuration (by default `/etc/gitea/conf/app.ini`):
+
+```ini
+[storage]
+STORAGE_TYPE=minio
+MINIO_ENDPOINT=localhost:3900
+MINIO_ACCESS_KEY_ID=GKxxx
+MINIO_SECRET_ACCESS_KEY=xxxx
+MINIO_BUCKET=gitea
+MINIO_LOCATION=garage
+MINIO_USE_SSL=false
+```
+
+You can also pass this configuration through environment variables:
+
+```bash
+GITEA__storage__STORAGE_TYPE=minio
+GITEA__storage__MINIO_ENDPOINT=localhost:3900
+GITEA__storage__MINIO_ACCESS_KEY_ID=GKxxx
+GITEA__storage__MINIO_SECRET_ACCESS_KEY=xxxx
+GITEA__storage__MINIO_BUCKET=gitea
+GITEA__storage__MINIO_LOCATION=garage
+GITEA__storage__MINIO_USE_SSL=false
+```
+
+Then restart your gitea instance and try to upload a custom avatar.
+If it worked, you should see some content in your gitea bucket (you must configure your `aws` command before):
+
+```
+$ aws s3 ls s3://gitea/avatars/
+2021-11-10 12:35:47     190034 616ba79ae2b84f565c33d72c2ec50861
+```
+
+
+*External link:* [Gitea Documentation > Configuration Cheat Sheet](https://docs.gitea.io/en-us/config-cheat-sheet/)
+
+## Docker
+
+Create a bucket and a key for your docker registry, then create `config.yml` with the following content:
+
+```yml
+version: 0.1
+http:
+  addr: 0.0.0.0:5000
+  secret: asecretforlocaldevelopment
+  debug:
+    addr: localhost:5001
+storage:
+  s3:
+    accesskey: GKxxxx
+    secretkey: yyyyy
+    region: garage
+    regionendpoint: http://localhost:3900
+    bucket: docker
+    secure: false
+    v4auth: true
+    rootdirectory: /
+```
+
+Replace the `accesskey`, `secretkey`, `bucket`, `regionendpoint` and `secure` values by the one fitting your deployment.
+
+Then simply run the docker registry:
+
+```bash
+docker run \
+  --net=host \
+  -v `pwd`/config.yml:/etc/docker/registry/config.yml \
+  registry:2
+```
+
+*We started a plain text registry but docker clients require encrypted registries. You must either [setup TLS](https://docs.docker.com/registry/deploying/#run-an-externally-accessible-registry) on your registry or add `--insecure-registry=localhost:5000` to your docker daemon parameters.*
+
+
+*External link:* [Docker Documentation > Registry storage drivers > S3 storage driver](https://docs.docker.com/registry/storage-drivers/s3/)
+
+## Nix
+
+Nix has no repository in its terminology: instead, it breaks down this concept in 2 parts: binary cache and channel.
+
+**A channel** is a set of `.nix` definitions that generate definitions for all the software you want to serve.
+
+Because we do not want all our clients to compile all these derivations by themselves,
+we can compile them once and then serve them as part of our **binary cache**.
+
+It is possible to use a **binary cache** without a channel, you only need to serve your nix definitions
+through another support, like a git repository.
+
+As a first step, we will need to create a bucket on Garage and enabling website access on it:
+
+```bash
+garage key new --name nix-key
+garage bucket create nix.example.com
+garage bucket allow nix.example.com --read --write --key nix-key
+garage bucket website nix.example.com --allow
+```
+
+If you need more information about exposing buckets as websites on Garage,
+check [Exposing buckets as websites](/cookbook/exposing_websites.html)
+ and [Configuring a reverse proxy](/cookbook/reverse_proxy.html).
+
+Next, we want to check that our bucket works:
+
+```bash
+echo nix repo > /tmp/index.html
+mc cp /tmp/index.html garage/nix/
+rm /tmp/index.html
+
+curl https://nix.example.com
+# output: nix repo
+```
+
+### Binary cache
+
+To serve binaries as part of your cache, you need to sign them with a key specific to nix.
+You can generate the keypair as follow: 
+
+```bash
+nix-store --generate-binary-cache-key <name> cache-priv-key.pem cache-pub-key.pem
+```
+
+You can then manually sign the packages of your store with the following command:
+
+```bash
+nix sign-paths --all -k cache-priv-key.pem
+```
+
+Setting a key in `nix.conf` will do the signature at build time automatically without additional commands.
+Edit the `nix.conf` of your builder:
+
+```toml
+secret-key-files = /etc/nix/cache-priv-key.pem
+```
+
+Now that your content is signed, you can copy a derivation to your cache.
+For example, if you want to copy a specific derivation of your store:
+
+```bash
+nix copy /nix/store/wadmyilr414n7bimxysbny876i2vlm5r-bash-5.1-p8 --to 's3://nix?endpoint=garage.example.com&region=garage'
+```
+
+*Note that if you have not signed your packages, you can append to the end of your S3 URL `&secret-key=/etc/nix/cache-priv-key.pem`.*
+
+Sometimes you don't want to hardcode this store path in your script.
+Let suppose that you are working on a codebase that you build with `nix-build`, you can then run:
+
+```bash
+nix copy $(nix-build) --to 's3://nix?endpoint=garage.example.com&region=garage'
+```
+
+*This command works because the only thing that `nix-build` outputs on stdout is the paths of the built derivations in your nix store.*
+
+You can include your derivation dependencies:
+
+```bash
+nix copy $(nix-store -qR $(nix-build)) --to 's3://nix?endpoint=garage.example.com&region=garage'
+```
+
+Now, your binary cache stores your derivation and all its dependencies.
+Just inform your users that they must update their `nix.conf` file with the following lines:
+
+```toml
+substituters = https://cache.nixos.org https://nix.example.com
+trusted-public-keys = cache.nixos.org-1:6NCHdD59X431o0gWypbMrAURkbJ16ZPMQFGspcDShjY= nix.example.com:eTGL6kvaQn6cDR/F9lDYUIP9nCVR/kkshYfLDJf1yKs=
+```
+
+*You must re-add cache.nixorg.org because redeclaring these keys override the previous configuration instead of extending it.*
+
+Now, when your clients will run `nix-build` or any command that generates a derivation for which a hash is already present
+on the binary cache, the client will download the result from the cache instead of compiling it, saving lot of time and CPU!
+
+
+### Channels
+
+Channels additionnaly serve Nix definitions, ie. a `.nix` file referencing
+all the derivations you want to serve.
+
+## Gitlab
+
+*External link:* [Gitlab Documentation > Object storage](https://docs.gitlab.com/ee/administration/object_storage.html)
+
+

doc/book/src/connect/websites.md

@@ -0,0 +1,83 @@
+# Websites (Hugo, Jekyll, Publii...)
+
+Garage is also suitable to host static websites.
+While they can be deployed with traditional CLI tools, some static website generators have integrated options to ease your workflow.
+
+| Name | Status | Note |
+|------|--------|------|
+| [Hugo](#hugo)     | ✅       | Publishing logic is integrated in the tool  |
+| [Publii](#publii)     | ✅       | Require a correctly configured s3 vhost endpoint     |
+| [Generic Static Site Generator](#generic-static-site-generator)     | ✅        |  Works for Jekyll, Zola, Gatsby, Pelican, etc.    |
+
+## Hugo
+
+Add to your `config.toml` the following section:
+
+```toml
+[[deployment.targets]]
+ URL = "s3://<bucket>?endpoint=<endpoint>&disableSSL=<bool>&s3ForcePathStyle=true&region=garage"
+```
+
+For example:
+
+```toml
+[[deployment.targets]]
+ URL = "s3://my-blog?endpoint=localhost:9000&disableSSL=true&s3ForcePathStyle=true&region=garage"
+```
+
+Then inform hugo of your credentials:
+
+```bash
+export AWS_ACCESS_KEY_ID=GKxxx
+export AWS_SECRET_ACCESS_KEY=xxx
+```
+
+And finally build and deploy your website:
+
+```bsh
+hugo
+hugo deploy
+```
+
+*External links:*
+  - [gocloud.dev > aws > Supported URL parameters](https://pkg.go.dev/gocloud.dev/aws?utm_source=godoc#ConfigFromURLParams)
+  - [Hugo Documentation > hugo deploy](https://gohugo.io/hosting-and-deployment/hugo-deploy/)
+
+## Publii
+
+[![A screenshot of Publii's GUI](./publii.png)](./publii.png)
+
+Deploying a website to Garage from Publii is natively supported.
+First, make sure that your Garage administrator allowed and configured Garage to support vhost access style.
+We also suppose that your bucket ("my-bucket") and key is already created and configured.
+
+Then, from the left menu, click on server. Choose "S3" as the protocol.
+In the configuration window, enter:
+  - Your finale website URL (eg. "http://my-bucket.web.garage.localhost:3902")
+  - Tick "Use a custom S3 provider"
+  - Set the S3 endpoint, (eg. "http://s3.garage.localhost:3900")
+  - Then put your access key (eg. "GK..."), your secret key, and your bucket (eg. "my-bucket")
+  - And hit the button "Save settings"
+
+Now, each time you want to publish your website from Publii, just hit the bottom left button "Sync your website"!
+
+
+
+## Generic Static Site Generator
+
+Some tools do not support sending to a S3 backend but output a compiled folder on your system.
+We can then use any CLI tool to upload this content to our S3 target.
+
+First, start by [configuring minio client](cli.html#minio-client-recommended).
+
+Then build your website (example for jekyll):
+
+```bash
+jekyll build
+```
+
+And copy its output folder (`_site` for Jekyll) on S3:
+
+```bash
+mc mirror --overwrite _site garage/my-site
+```

doc/book/src/cookbook/clients.md

@@ -1,105 +0,0 @@
-# Configuring S3 clients to interact with Garage
-
-To configure an S3 client to interact with Garage, you will need the following
-parameters:
-
-- An **API endpoint**: this corresponds to the HTTP or HTTPS address
-  used to contact the Garage server. When runing Garage locally this will usually
-  be `http://127.0.0.1:3900`. In a real-world setting, you would usually have a reverse-proxy
-  that adds TLS support and makes your Garage server available under a public hostname
-  such as `https://garage.example.com`.
-
-- An **API access key** and its associated **secret key**. These usually look something
-  like this: `GK3515373e4c851ebaad366558` (access key),
-  `7d37d093435a41f2aab8f13c19ba067d9776c90215f56614adad6ece597dbb34` (secret key).
-  These keys are created and managed using the `garage` CLI, as explained in the
-  [quick start](../quick_start/index.md) guide.
-
-Most S3 clients can be configured easily with these parameters,
-provided that you follow the following guidelines:
-
-- **Force path style:** Garage does not support DNS-style buckets, which are now by default
-  on Amazon S3. Instead, Garage uses the legacy path-style bucket addressing.
-  Remember to configure your client to acknowledge this fact.
-
-- **Configuring the S3 region:** Garage requires your client to talk to the correct "S3 region",
-  which is set in the configuration file. This is often set just to `garage`.
-  If this is not configured explicitly, clients usually try to talk to region `us-east-1`.
-  Garage should normally redirect your client to the correct region,
-  but in case your client does not support this you might have to configure it manually.
-
-We will now provide example configurations for the most common S3 clients.
-
-## AWS CLI
-
-Export the following environment variables:
-
-```bash
-export AWS_ACCESS_KEY_ID=<access key>
-export AWS_SECRET_ACCESS_KEY=<secret key>
-export AWS_DEFAULT_REGION=<region>
-```
-
-Now invoke `aws` as follows:
-
-```bash
-aws --endpoint-url <endpoint> s3 <command...>
-```
-
-For instance: `aws --endpoint-url http://127.0.0.1:3901 s3 ls s3://my-bucket/`.
-
-## Minio client
-
-Use the following command to set an "alias", i.e. define a new S3 server to be
-used by the Minio client:
-
-```bash
-mc alias set \
-  garage \
-  <endpoint> \
-  <access key> \
-  <secret key> \
-  --api S3v4
-```
-
-Remember that `mc` is sometimes called `mcli` (such as on Arch Linux), to avoid conflicts
-with the Midnight Commander.
-
-
-## `rclone`
-
-`rclone` can be configured using the interactive assistant invoked using `rclone configure`.
-
-You can also configure `rclone` by writing directly its configuration file.
-Here is a template `rclone.ini` configuration file:
-
-```ini
-[garage]
-type = s3
-provider = Other
-env_auth = false
-access_key_id = <access key>
-secret_access_key = <secret key>
-region = <region>
-endpoint = <endpoint>
-force_path_style = true
-acl = private
-bucket_acl = private
-```
-
-## Cyberduck
-
-TODO
-
-## `s3cmd`
-
-Here is a template for the `s3cmd.cfg` file to talk with Garage:
-
-```ini
-[default]
-access_key = <access key>
-secret_key = <secret key>
-host_base = <endpoint without http(s)://>
-host_bucket = <same as host_base>
-use_https = False | True
-```

doc/book/src/cookbook/exposing_websites.md

@@ -0,0 +1,48 @@
+# Exposing buckets as websites
+
+You can expose your bucket as a website with this simple command:
+
+```bash
+garage bucket website --allow my-website
+```
+
+Now it will be **publicly** exposed on the web endpoint (by default listening on port 3902).
+
+Our website serving logic is as follow:
+  - Supports only static websites (no support for PHP or other languages)
+  - Does not support directory listing
+  - The index is defined in your `garage.toml`. ([ref](/reference_manual/configuration.html#index))
+
+Now we need to infer the URL of your website through your bucket name.
+Let assume:
+  - we set `root_domain = ".web.example.com"` in `garage.toml` ([ref](/reference_manual/configuration.html#root_domain))
+  - our bucket name is `garagehq.deuxfleurs.fr`.
+
+Our bucket will be served if the Host field matches one of these 2 values (the port is ignored):
+
+  - `garagehq.deuxfleurs.fr.web.example.com`: you can dedicate a subdomain to your users (here `web.example.com`).
+
+  - `garagehq.deuxfleurs.fr`: your users can bring their own domain name, they just need to point them to your Garage cluster.
+
+You can try this logic locally, without configuring any DNS, thanks to `curl`:
+
+```bash
+# prepare your test
+echo hello world > /tmp/index.html
+mc cp /tmp/index.html garage/garagehq.deuxfleurs.fr
+
+curl -H 'Host: garagehq.deuxfleurs.fr' http://localhost:3902
+# should print "hello world"
+
+curl -H 'Host: garagehq.deuxfleurs.fr.web.example.com' http://localhost:3902
+# should also print "hello world"
+```
+
+Now that you understand how website logic works on Garage, you can:
+
+ - make the website endpoint listens on port 80 (instead of 3902)
+ - use iptables to redirect the port 80 to the port 3902:  
+   `iptables -t nat -A PREROUTING -p tcp -dport 80 -j REDIRECT -to-port 3902`
+ - or configure a [reverse proxy](reverse_proxy.html) in front of Garage to add TLS (HTTPS), CORS support, etc.
+
+You can also take a look at [Website Integration](/connect/websites.html) to see how you can add Garage to your workflow.

doc/book/src/cookbook/gateways.md

@@ -0,0 +1,39 @@
+# Gateways
+
+Gateways allow you to expose Garage endpoints (S3 API and websites) without storing data on the node.
+
+## Benefits
+
+You can configure Garage as a gateway on all nodes that will consume your S3 API, it will provide you the following benefits:
+
+  - **It removes 1 or 2 network RTT.** Instead of (querying your reverse proxy then) querying a random node of the cluster that will forward your request to the nodes effectively storing the data, your local gateway will directly knows which node to query.
+
+  - **It eases server management.** Instead of tracking in your reverse proxy and DNS what are the current Garage nodes, your gateway being part of the cluster keeps this information for you. In your software, you will always specify `http://localhost:3900`.
+
+  - **It simplifies security.** Instead of having to maintain and renew a TLS certificate, you leverage the Secret Handshake protocol we use for our cluster. The S3 API protocol will be in plain text but limited to your local machine.
+
+## Limitations
+
+Currently it will not work with minio client. Follow issue [#64](https://git.deuxfleurs.fr/Deuxfleurs/garage/issues/64) for more information.
+
+## Spawn a Gateway
+
+The instructions are similar to a regular node, the only option that is different is while configuring the node, you must set the `--gateway` parameter:
+
+```bash
+garage layout assign --gateway --tag gw1 <node_id>
+garage layout show    # review the changes you are making
+garage layout apply   # once satisfied, apply the changes
+```
+
+Then use `http://localhost:3900` when a S3 endpoint is required:
+
+```bash
+aws --endpoint-url http://127.0.0.1:3900 s3 ls
+```
+
+If a newly added gateway node seems to not be working, do a full table resync to ensure that bucket and key list are correctly propagated:
+
+```bash
+garage repair -a --yes tables
+```

doc/book/src/cookbook/index.md

@@ -4,22 +4,23 @@ A cookbook, when you cook, is a collection of recipes.
 Similarly, Garage's cookbook contains a collection of recipes that are known to works well!
 This chapter could also be referred as "Tutorials" or "Best practices".
 
-- **[Deploying Garage](real_world.md):** This page will walk you through all of the necessary
+- **[Multi-node deployment](real_world.md):** This page will walk you through all of the necessary
   steps to deploy Garage in a real-world setting.
 
-- **[Configuring S3 clients](clients.md):** This page will explain how to configure
-  popular S3 clients to interact with a Garage server.
-
-- **[Hosting a website](website.md):** This page explains how to use Garage
-  to host a static website.
-
-- **[Recovering from failures](recovering.md):** Garage's first selling point is resilience
-  to hardware failures. This section explains how to recover from such a failure in the
-  best possible way.
-
 - **[Building from source](from_source.md):** This page explains how to build Garage from
   source in case a binary is not provided for your architecture, or if you want to
   hack with us!
 
-- **[Starting with Systemd](from_source.md):** This page explains how to run Garage
+- **[Integration with Systemd](systemd.md):** This page explains how to run Garage
   as a Systemd service (instead of as a Docker container).
+
+- **[Configuring a gateway node](gateways.md):** This page explains how to run a gateway node in a Garage cluster, i.e. a Garage node that doesn't store data but accelerates access to data present on the other nodes.
+
+- **[Hosting a website](exposing_websites.md):** This page explains how to use Garage
+  to host a static website.
+
+- **[Configuring a reverse-proxy](reverse_proxy.md):** This page explains how to configure a reverse-proxy to add TLS support to your S3 api endpoint.
+
+- **[Recovering from failures](recovering.md):** Garage's first selling point is resilience
+  to hardware failures. This section explains how to recover from such a failure in the
+  best possible way.

doc/book/src/cookbook/real_world.md

@@ -11,15 +11,12 @@ to get familiar with Garage's command line and usage patterns.
 
 ## Prerequisites
 
-To run a real-world deployment, make sure you the following conditions are met:
+To run a real-world deployment, make sure the following conditions are met:
 
 - You have at least three machines with sufficient storage space available.
 
 - Each machine has a public IP address which is reachable by other machines.
-  Running behind a NAT is possible, but having several Garage nodes behind a single NAT
-  is slightly more involved as each will have to have a different RPC port number
-  (the local port number of a node must be the same as the port number exposed publicly
-  by the NAT).
+  Running behind a NAT is likely to be possible but hasn't been tested for the latest version (TODO).
 
 - Ideally, each machine should have a SSD available in addition to the HDD you are dedicating
   to Garage. This will allow for faster access to metadata and has the potential
@@ -44,45 +41,23 @@ For our example, we will suppose the following infrastructure with IPv6 connecti
 
 ## Get a Docker image
 
-Our docker image is currently named `lxpz/garage_amd64` and is stored on the [Docker Hub](https://hub.docker.com/r/lxpz/garage_amd64/tags?page=1&ordering=last_updated).
-We encourage you to use a fixed tag (eg. `v0.3.0`) and not the `latest` tag.
-For this example, we will use the latest published version at the time of the writing which is `v0.3.0` but it's up to you
-to check [the most recent versions on the Docker Hub](https://hub.docker.com/r/lxpz/garage_amd64/tags?page=1&ordering=last_updated).
+Our docker image is currently named `dxflrs/amd64_garage` and is stored on the [Docker Hub](https://hub.docker.com/r/dxflrs/amd64_garage/tags?page=1&ordering=last_updated).
+We encourage you to use a fixed tag (eg. `v0.4.0`) and not the `latest` tag.
+For this example, we will use the latest published version at the time of the writing which is `v0.4.0` but it's up to you
+to check [the most recent versions on the Docker Hub](https://hub.docker.com/r/dxflrs/amd64_garage/tags?page=1&ordering=last_updated).
 
 For example:
 
 ```
-sudo docker pull lxpz/garage_amd64:v0.3.0
+sudo docker pull dxflrs/amd64_garage:v0.4.0
 ```
 
-
-## Generating TLS certificates
-
-You first need to generate TLS certificates to encrypt traffic between Garage nodes
-(reffered to as RPC traffic).
-
-To generate your TLS certificates, run on your machine:
-
-```
-wget https://git.deuxfleurs.fr/Deuxfleurs/garage/raw/branch/main/genkeys.sh
-chmod +x genkeys.sh
-./genkeys.sh
-```
-
-It will creates a folder named `pki/` containing the keys that you will used for the cluster.
-These files will have to be copied to all of your cluster nodes, as explained below.
-
-
 ## Deploying and configuring Garage
 
 On each machine, we will have a similar setup,
 especially you must consider the following folders/files:
 
-- `/etc/garage/garage.toml`: Garage daemon's configuration (see below)
-
-- `/etc/garage/pki/`: Folder containing Garage certificates,
-  must be generated on your computer and copied on the servers.
-  Only the files `garage-ca.crt`, `garage.crt` and `garage.key` are necessary.
+- `/etc/garage.toml`: Garage daemon's configuration (see below)
 
 - `/var/lib/garage/meta/`: Folder containing Garage's metadata,
   put this folder on a SSD if possible
@@ -91,7 +66,7 @@ especially you must consider the following folders/files:
   this folder will be your main data storage and must be on a large storage (e.g. large HDD)
 
 
-A valid `/etc/garage/garage.toml` for our cluster would be:
+A valid `/etc/garage/garage.toml` for our cluster would look as follows:
 
 ```toml
 metadata_dir = "/var/lib/garage/meta"
@@ -99,23 +74,18 @@ data_dir = "/var/lib/garage/data"
 
 replication_mode = "3"
 
+compression_level = 2
+
 rpc_bind_addr = "[::]:3901"
+rpc_public_addr = "<this node's public IP>:3901"
+rpc_secret = "<RPC secret>"
 
-bootstrap_peers = [
-  "[fc00:1::1]:3901",
-  "[fc00:1::2]:3901",
-  "[fc00:B::1]:3901",
-  "[fc00:F::1]:3901",
-]
-
-[rpc_tls]
-ca_cert = "/etc/garage/pki/garage-ca.crt"
-node_cert = "/etc/garage/pki/garage.crt"
-node_key = "/etc/garage/pki/garage.key"
+bootstrap_peers = []
 
 [s3_api]
 s3_region = "garage"
 api_bind_addr = "[::]:3900"
+root_domain = ".s3.garage"
 
 [s3_web]
 bind_addr = "[::]:3902"
@@ -123,11 +93,14 @@ root_domain = ".web.garage"
 index = "index.html"
 ```
 
-Please make sure to change `bootstrap_peers` to **your** IP addresses!
+Check the following for your configuration files:
 
-Check the [configuration file reference documentation](../reference_manual/configuration.md)
-to learn more about all available configuration options.
+- Make sure `rpc_public_addr` contains the public IP address of the node you are configuring.
+  This parameter is optional but recommended: if your nodes have trouble communicating with
+  one another, consider adding it.
 
+- Make sure `rpc_secret` is the same value on all nodes. It should be a 32-bytes hex-encoded secret key.
+  You can generate such a key with `openssl rand -hex 32`.
 
 ## Starting Garage using Docker
 
@@ -139,11 +112,10 @@ docker run \
   --name garaged \
   --restart always \
   --network host \
-  -v /etc/garage/pki:/etc/garage/pki \
-  -v /etc/garage/garage.toml:/garage/garage.toml \
+  -v /etc/garage.toml:/etc/garage.toml \
   -v /var/lib/garage/meta:/var/lib/garage/meta \
   -v /var/lib/garage/data:/var/lib/garage/data \
-  lxpz/garage_amd64:v0.3.0
+  lxpz/garage_amd64:v0.4.0
 ```
 
 It should be restarted automatically at each reboot.
@@ -155,101 +127,104 @@ but please check the relase notes before doing so!
 To upgrade, simply stop and remove this container and
 start again the command with a new version of Garage.
 
-
 ## Controling the daemon
 
 The `garage` binary has two purposes:
-  - it acts as a daemon when launched with `garage server ...`
+  - it acts as a daemon when launched with `garage server`
   - it acts as a control tool for the daemon when launched with any other command
 
-In this section, we will see how to use the `garage` binary as a control tool for the daemon we just started.
-You first need to get a shell having access to this binary. For instance, enter the Docker container with:
+Ensure an appropriate `garage` binary (the same version as your Docker image) is available in your path.
+If your configuration file is at `/etc/garage.toml`, the `garage` binary should work with no further change.
+
+You can test your `garage` CLI utility by running a simple command such as:
 
 ```bash
-sudo docker exec -ti garaged bash
+garage status
 ```
 
-You will now have a shell where the Garage binary is available as `/garage/garage`
-
-*You can also install the binary on your machine to remotely control the cluster.*
-
-## Talk to the daemon and create an alias
-
-`garage` requires 4 options to talk with the daemon:
+At this point, nodes are not yet talking to one another.
+Your output should therefore look like follows:
 
 ```
---ca-cert <ca-cert>            
---client-cert <client-cert>    
---client-key <client-key>      
--h, --rpc-host <rpc-host>
+Mercury$ garage status
+==== HEALTHY NODES ====
+ID                  Hostname  Address           Tag                   Zone  Capacity
+563e1ac825ee3323…   Mercury   [fc00:1::1]:3901  NO ROLE ASSIGNED
 ```
 
-The 3 first ones are certificates and keys needed by TLS, the last one is simply the address of Garage's RPC endpoint.
 
-If you are invoking `garage` from a server node directly, you do not need to set `--rpc-host`
-as the default value `127.0.0.1:3901` will allow it to contact Garage correctly.
+## Connecting nodes together
 
-To avoid typing the 3 first options each time we want to run a command,
-you can use the following alias:
+When your Garage nodes first start, they will generate a local node identifier
+(based on a public/private key pair).
+
+To obtain the node identifier of a node, once it is generated,
+run `garage node id`.
+This will print keys as follows:
 
 ```bash
-alias garagectl='/garage/garage \
-  --ca-cert /etc/garage/pki/garage-ca.crt \
-  --client-cert /etc/garage/pki/garage.crt \
-  --client-key /etc/garage/pki/garage.key'
+Mercury$ garage node id
+563e1ac825ee3323aa441e72c26d1030d6d4414aeb3dd25287c531e7fc2bc95d@[fc00:1::1]:3901
+
+Venus$ garage node id
+86f0f26ae4afbd59aaf9cfb059eefac844951efd5b8caeec0d53f4ed6c85f332@[fc00:1::2]:3901
+
+etc.
 ```
 
-You can now use all of the commands presented in the [quick start guide](../quick_start/index.md),
-simply replace occurences of `garage` by `garagectl`.
+You can then instruct nodes to connect to one another as follows:
 
-#### Test the alias
-
-You can test your alias by running a simple command such as:
-
-```
-garagectl status
+```bash
+# Instruct Venus to connect to Mercury (this will establish communication both ways)
+Venus$ garage node connect 563e1ac825ee3323aa441e72c26d1030d6d4414aeb3dd25287c531e7fc2bc95d@[fc00:1::1]:3901
 ```
 
-You should get something like that as result:
+You don't nead to instruct all node to connect to all other nodes:
+nodes will discover one another transitively.
+
+Now if your run `garage status` on any node, you should have an output that looks as follows:
 
 ```
-Healthy nodes:
-8781c50c410a41b3…	Mercury	[fc00:1::1]:3901	UNCONFIGURED/REMOVED
-2a638ed6c775b69a…	Venus	[fc00:1::2]:3901	UNCONFIGURED/REMOVED
-68143d720f20c89d…	Earth	[fc00:B::1]:3901	UNCONFIGURED/REMOVED
-212f7572f0c89da9…	Mars	[fc00:F::1]:3901	UNCONFIGURED/REMOVED
+==== HEALTHY NODES ====
+ID                  Hostname  Address           Tag                   Zone  Capacity
+563e1ac825ee3323…   Mercury   [fc00:1::1]:3901  NO ROLE ASSIGNED
+86f0f26ae4afbd59…   Venus     [fc00:1::2]:3901  NO ROLE ASSIGNED
+68143d720f20c89d…   Earth     [fc00:B::1]:3901  NO ROLE ASSIGNED
+212f7572f0c89da9…   Mars      [fc00:F::1]:3901  NO ROLE ASSIGNED
 ```
 
-
-## Configuring a cluster
+## Creating a cluster layout
 
 We will now inform Garage of the disk space available on each node of the cluster
 as well as the zone (e.g. datacenter) in which each machine is located.
+This information is called the **cluster layout** and consists
+of a role that is assigned to each active cluster node.
 
-For our example, we will suppose we have the following infrastructure (Capacity, Identifier and Datacenter are specific values to Garage described in the following):
+For our example, we will suppose we have the following infrastructure
+(Capacity, Identifier and Zone are specific values to Garage described in the following):
 
 | Location | Name    | Disk Space | `Capacity` | `Identifier` | `Zone` |
 |----------|---------|------------|------------|--------------|--------------|
-| Paris    | Mercury | 1 To       | `2`        | `8781c5`     | `par1`       |
-| Paris    | Venus   | 2 To       | `4`        | `2a638e`     | `par1`       |
-| London   | Earth   | 2 To       | `4`        | `68143d`     | `lon1`       |
-| Brussels | Mars    | 1.5 To     | `3`        | `212f75`     | `bru1`       |
+| Paris    | Mercury | 1 To       | `10`       | `563e`     | `par1`       |
+| Paris    | Venus   | 2 To       | `20`       | `86f0`     | `par1`       |
+| London   | Earth   | 2 To       | `20`       | `6814`     | `lon1`       |
+| Brussels | Mars    | 1.5 To     | `15`       | `212f`     | `bru1`       |
 
 #### Node identifiers
 
 After its first launch, Garage generates a random and unique identifier for each nodes, such as:
 
 ```
-8781c50c410a41b363167e9d49cc468b6b9e4449b6577b64f15a249a149bdcbc
+563e1ac825ee3323aa441e72c26d1030d6d4414aeb3dd25287c531e7fc2bc95d
 ```
 
-Often a shorter form can be used, containing only the beginning of the identifier, like `8781c5`,
+Often a shorter form can be used, containing only the beginning of the identifier, like `563e`,
 which identifies the server "Mercury" located in "Paris" according to our previous table.
 
 The most simple way to match an identifier to a node is to run:
 
 ```
-garagectl status
+garage status
 ```
 
 It will display the IP address associated with each node;
@@ -268,13 +243,9 @@ in order to provide high availability despite failure of a zone.
 
 Garage reasons on an abstract metric about disk storage that is named the *capacity* of a node.
 The capacity configured in Garage must be proportional to the disk space dedicated to the node.
-Due to the way the Garage allocation algorithm works, capacity values must
-be **integers**, and must be **as small as possible**, for instance with
-1 representing the size of your smallest server.
 
-Here we chose that 1 unit of capacity = 0.5 To, so that we can express servers of size
-1 To and 2 To, as wel as the intermediate size 1.5 To, with the integer values 2, 4 and
-3 respectively (see table above).
+Capacity values must be **integers** but can be given any signification.
+Here we chose that 1 unit of capacity = 100 GB.
 
 Note that the amount of data stored by Garage on each server may not be strictly proportional to
 its capacity value, as Garage will priorize having 3 copies of data in different zones,
@@ -286,20 +257,36 @@ have 66% chance of being stored by Venus and 33% chance of being stored by Mercu
 
 Given the information above, we will configure our cluster as follow:
 
+```bash
+garage layout assign -z par1 -c 10 -t mercury 563e
+garage layout assign -z par1 -c 20 -t venus 86f0
+garage layout assign -z lon1 -c 20 -t earth 6814
+garage layout assign -z bru1 -c 15 -t mars 212f
 ```
-garagectl node configure -z par1 -c 2 -t mercury 8781c5
-garagectl node configure -z par1 -c 4 -t venus 2a638e
-garagectl node configure -z lon1 -c 4 -t earth 68143d
-garagectl node configure -z bru1 -c 3 -t mars 212f75
+
+At this point, the changes in the cluster layout have not yet been applied.
+To show the new layout that will be applied, call:
+
+```bash
+garage layout show
 ```
 
+Once you are satisfied with your new layout, apply it with:
+
+```bash
+garage layout apply
+```
+
+**WARNING:** if you want to use the layout modification commands in a script,
+make sure to read [this page](/reference_manual/layout.html) first.
+
 
 ## Using your Garage cluster
 
-Creating buckets and managing keys is done using the `garagectl` CLI,
+Creating buckets and managing keys is done using the `garage` CLI,
 and is covered in the [quick start guide](../quick_start/index.md).
 Remember also that the CLI is self-documented thanks to the `--help` flag and
 the `help` subcommand (e.g. `garage help`, `garage key --help`).
 
-Configuring an S3 client to interact with Garage is covered
-[in the next section](clients.md).
+Configuring S3-compatible applicatiosn to interact with Garage
+is covered in the [Integrations](/connect/index.html) section.

doc/book/src/cookbook/recovering.md

@@ -28,8 +28,10 @@ and you should instead use one of the methods detailed in the next sections.
 
 Removing a node is done with the following command:
 
-```
-garage node remove --yes <node_id>
+```bash
+garage layout remove <node_id>
+garage layout show    # review the changes you are making
+garage layout apply   # once satisfied, apply the changes
 ```
 
 (you can get the `node_id` of the failed node by running `garage status`)
@@ -50,7 +52,7 @@ We just need to tell Garage to get back all the data blocks and store them on th
 First, set up a new HDD to store Garage's data directory on the failed node, and restart Garage using
 the existing configuration.  Then, run:
 
-```
+```bash
 garage repair -a --yes blocks
 ```
 
@@ -58,7 +60,7 @@ This will re-synchronize blocks of data that are missing to the new HDD, reading
 
 You can check on the advancement of this process by doing the following command: 
 
-```
+```bash
 garage stats -a
 ```
 
@@ -94,9 +96,11 @@ The ID of the lost node should be shown in `garage status` in the section for di
 
 Then, replace the broken node by the new one, using:
 
-```
-garage node configure --replace <old_node_id> \
-		-c <capacity> -z <zone> -t <node_tag> <new_node_id>
+```bash
+garage layout assign <new_node_id> --replace <old_node_id> \
+		-c <capacity> -z <zone> -t <node_tag>
+garage layout show    # review the changes you are making
+garage layout apply   # once satisfied, apply the changes
 ```
 
 Garage will then start synchronizing all required data on the new node.

doc/book/src/cookbook/reverse_proxy.md

@@ -0,0 +1,142 @@
+# Configuring a reverse proxy
+
+The main reason to add a reverse proxy in front of Garage is to provide TLS to your users and serve multiple web services on port 443.
+
+In production you will likely need your certificates signed by a certificate authority.
+The most automated way is to use a provider supporting the [ACME protocol](https://datatracker.ietf.org/doc/html/rfc8555) 
+such as [Let's Encrypt](https://letsencrypt.org/), [ZeroSSL](https://zerossl.com/) or [Buypass Go SSL](https://www.buypass.com/ssl/products/acme).
+
+If you are only testing Garage, you can generate a self-signed certificate to follow the documentation:
+
+```bash
+openssl req \
+    -new \
+    -x509 \
+    -keyout /tmp/garage.key \
+    -out /tmp/garage.crt \
+    -nodes  \
+    -subj "/C=XX/ST=XX/L=XX/O=XX/OU=XX/CN=localhost/emailAddress=X@X.XX" \
+    -addext "subjectAltName = DNS:localhost, IP:127.0.0.1"
+
+cat /tmp/garage.key /tmp/garage.crt > /tmp/garage.pem
+```
+
+Be careful as you will need to allow self signed certificates in your client.
+For example, with minio, you must add the `--insecure` flag.
+An example:
+
+```bash
+mc ls --insecure garage/
+```
+
+## socat (only for testing purposes)
+
+If you want to test Garage with a TLS frontend, socat can do it for you in a single command:
+
+```bash
+socat \
+"openssl-listen:443,\
+reuseaddr,\
+fork,\
+verify=0,\
+cert=/tmp/garage.pem" \
+tcp4-connect:localhost:3900
+```
+
+## Nginx
+
+Nginx is a well-known reverse proxy suitable for production.
+We do the configuration in 3 steps: first we define the upstream blocks ("the backends")
+then we define the server blocks ("the frontends") for the S3 endpoint and finally for the web endpoint.
+
+The following configuration blocks can be all put in the same `/etc/nginx/sites-available/garage.conf`.
+To make your configuration active, run `ln -s /etc/nginx/sites-available/garage.conf /etc/nginx/sites-enabled/`.
+If you directly put the instructions in the root `nginx.conf`, keep in mind that these configurations must be enclosed inside a `http { }` block.
+
+And do not forget to reload nginx with `systemctl reload nginx` or `nginx -s reload`.
+
+### Exposing the S3 endpoints
+
+First, we need to tell to nginx how to access our Garage cluster.
+Because we have multiple nodes, we want to leverage all of them by spreading the load.
+In nginx, we can do that with the `upstream` directive.
+
+Then in a `server` directive, we define the vhosts, the TLS certificates and the proxy rule.
+
+A possible configuration:
+
+```nginx
+upstream s3_backend {
+  # if you have a garage instance locally
+  server 127.0.0.1:3900;
+  # you can also put your other instances
+  server 192.168.1.3:3900;
+  # domain names also work
+  server garage1.example.com:3900;
+  # you can assign weights if you have some servers 
+  # that are more powerful than others
+  server garage2.example.com:3900 weight=2;
+}
+
+server {
+  listen [::]:443 http2 ssl;
+
+  ssl_certificate     /tmp/garage.crt;
+  ssl_certificate_key /tmp/garage.key;
+
+  # You need multiple server names here:
+  #  - s3.garage.tld is used for path-based s3 requests
+  #  - *.s3.garage.tld is used for vhost-based s3 requests
+  server_name s3.garage.tld *.s3.garage.tld;
+
+  location / {
+    proxy_pass http://s3_backend;
+    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+    proxy_set_header Host $host;
+  }
+}
+```
+
+## Exposing the web endpoint
+
+To better understand the logic involved, you can refer to the [Exposing buckets as websites](/cookbook/exposing_websites.html) section.
+Otherwise, the configuration is very similar to the S3 endpoint.
+You must only adapt `upstream` with the web port instead of the s3 port and change the `server_name` and `proxy_pass` entry
+
+A possible configuration:
+
+
+```nginx
+upstream web_backend {
+  server 127.0.0.1:3902;
+  server 192.168.1.3:3902;
+  server garage1.example.com:3902;
+  server garage2.example.com:3902 weight=2;
+}
+
+server {
+  listen [::]:443 http2 ssl;
+
+  ssl_certificate     /tmp/garage.crt;
+  ssl_certificate_key /tmp/garage.key;
+
+  # You need multiple server names here:
+  #  - *.web.garage.tld is used for your users wanting a website without reserving a domain name
+  #  - example.com, my-site.tld, etc. are reserved domain name by your users that chose to host their website as a garage's bucket
+  server_name *.web.garage.tld example.com my-site.tld;
+
+  location / {
+    proxy_pass http://web_backend;
+    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+    proxy_set_header Host $host;
+  }
+}
+```
+
+## Apache httpd
+
+@TODO
+
+## Traefik
+
+@TODO

doc/book/src/cookbook/systemd.md

@@ -1,9 +1,14 @@
-# Starting Garage with systemd instead of Docker
+# Starting Garage with systemd
+
+We make some assumptions for this systemd deployment. 
+
+  - Your garage binary is located at `/usr/local/bin/garage`.
+
+  - Your configuration file is located at `/etc/garage.toml`.
+
+  - Your `garage.toml` must be set with  `metadata_dir=/var/lib/garage/meta` and `data_dir=/var/lib/garage/data`. This is mandatory to use `systemd` hardening feature [Dynamic User](https://0pointer.net/blog/dynamic-users-with-systemd.html). Note that in your host filesystem, Garage data will be held in `/var/lib/private/garage`.
 
-NOTE: This guide is incomplete. Typicall you would also want to create a separate
-Unix user to run Garage.
 
-Make sure you have the Garage binary installed on your system (see [quick start](../quick_start/index.md)), e.g. at `/usr/local/bin/garage`.
 
 Create a file named `/etc/systemd/system/garage.service`:
 
@@ -15,12 +20,18 @@ Wants=network-online.target
 
 [Service]
 Environment='RUST_LOG=garage=info' 'RUST_BACKTRACE=1'
-ExecStart=/usr/local/bin/garage server -c /etc/garage/garage.toml
+ExecStart=/usr/local/bin/garage server
+StateDirectory=garage
+DynamicUser=true
+ProtectHome=true
+NoNewPrivileges=true
 
 [Install]
 WantedBy=multi-user.target
 ```
 
+*A note on hardening: garage will be run as a non privileged user, its user id is dynamically allocated by systemd. It cannot access (read or write) home folders (/home, /root and /run/user), the rest of the filesystem can only be read but not written, only the path seen as /var/lib/garage is writable as seen by the service (mapped to /var/lib/private/garage on your host). Additionnaly, the process can not gain new privileges over time.*
+
 To start the service then automatically enable it at boot:
 
 ```bash

doc/book/src/design/benchmarks.md

@@ -0,0 +1,81 @@
+# Benchmarks
+
+With Garage, we wanted to build a software defined storage service that follow the [KISS principle](https://en.wikipedia.org/wiki/KISS_principle),
+ that is suitable for geo-distributed deployments and more generally that would work well for community hosting (like a Mastodon instance).
+
+In our benchmarks, we aim to quantify how Garage performs on these goals compared to the other available solutions.
+
+## Geo-distribution
+
+The main challenge in a geo-distributed setup is latency between nodes of the cluster.
+The more a user request will require intra-cluster requests to complete, the more its latency will increase.
+This is especially true for sequential requests: requests that must wait the result of another request to be sent.
+We designed Garage without consensus algorithms (eg. Paxos or Raft) to minimize the number of sequential and parallel requests.
+
+This serie of benchmarks quantifies the impact of this design choice.
+
+### On a simple simulated network
+
+We start with a controlled environment, all the instances are running on the same (powerful enough) machine.
+
+To control the network latency, we simulate the network with [mknet](https://git.deuxfleurs.fr/trinity-1686a/mknet) (a tool we developped, based on `tc` and the linux network stack).
+To mesure S3 endpoints latency, we use our own tool [s3lat](https://git.deuxfleurs.fr/quentin/s3lat/) to observe only the intra-cluster latency and not some contention on the nodes (CPU, RAM, disk I/O, network bandwidth, etc.).
+Compared to other benchmark tools, S3Lat sends only one (small) request at the same time and measures its latency.
+We selected 5 standard endpoints that are often in the critical path: ListBuckets, ListObjects, GetObject, PutObject and RemoveObject.
+
+In this first benchmark, we consider 5 instances that are located in a different place each. To simulate the distance, we configure mknet with a RTT between each node of 100 ms +/- 20 ms of jitter. We get the following graph, where the colored bars represent the mean latency while the error bars the minimum and maximum one:
+
+![Comparison of endpoints latency for minio and garage](./img/endpoint-latency.png)
+
+Compared to garage, minio latency drastically increases on 3 endpoints: GetObject, PutObject, RemoveObject.
+
+We suppose that these requests on minio make transactions over Raft, involving 4 sequential requests: 1) sending the message to the leader, 2) having the leader dispatch it to the other nodes, 3) waiting for the confirmation of followers and finally 4) commiting it. With our current configuration, one Raft transaction will take around 400 ms. GetObject seems to correlate to 1 transaction while PutObject and RemoveObject seems to correlate to 2 or 3. Reviewing minio code would be required to confirm this hypothesis.
+
+Conversely, garage uses an architecture similar to DynamoDB and never require global cluster coordination to answer a request.
+Instead, garage can always contact the right node in charge of the requested data, and can answer in as low as one request in the case of GetObject and PutObject. We also observed that Garage latency, while often lower to minio, is more dispersed: garage is still in beta and has not received any performance optimization yet.
+
+As a conclusion, Garage performs well in such setup while minio will be hard to use, especially for interactive use cases.
+
+### On a complex simulated network
+
+This time we consider a more heterogeneous network with 6 servers spread in 3 datacenter, giving us 2 servers per datacenters.
+We consider that intra-DC communications are now very cheap with a latency of 0.5ms and without any jitter.
+The inter-DC remains costly with the same value as before (100ms +/- 20ms of jitter).
+We plot a similar graph as before:
+
+![Comparison of endpoints latency for minio and garage with 6  nodes in 3 DC](./img/endpoint-latency-dc.png)
+
+This new graph is very similar to the one before, neither minio or garage seems to benefit from this new topology, but they also do not suffer from it.
+
+Considering garage, this is expected: nodes in the same DC are put in the same zone, and then data are spread on different zones for data resiliency and availaibility.
+Then, in the default mode, requesting data requires to query at least 2 zones to be sure that we have the most up to date information.
+These requests will involve at least one inter-DC communication.
+In other words, we prioritize data availability and synchronization over raw performances.
+
+Minio's case is a bit different as by default a minio cluster is not location aware, so we can't explain its performances through location awareness.
+*We know that minio has a multi site mode but it is definitely not a first class citizen: data are asynchronously replicated from one minio cluster to another.*
+We suppose that, due to the consensus, for many of its requests minio will wait for a response of the majority of the server, also involving inter-DC communications.
+
+As a conclusion, our new topology did not influence garage or minio performances, confirming that in presence of latency, garage is the best fit.
+
+### On a real world deployment
+
+*TODO*
+
+
+## Performance stability
+
+A storage cluster will encounter different scenario over its life, many of them will not be predictable.
+In this context, we argue that, more than peak performances, we should seek predictable and stable performances to ensure data availability.
+
+### Reference
+
+*TODO*
+
+### On a degraded cluster
+
+*TODO*
+
+### At scale
+
+*TODO*

doc/book/src/design/goals.md

@@ -0,0 +1,53 @@
+# Goals and use cases
+
+## Goals and non-goals
+
+Garage is a lightweight geo-distributed data store that implements the
+[Amazon S3](https://docs.aws.amazon.com/AmazonS3/latest/API/Welcome.html)
+object storage protocole. It enables applications to store large blobs such
+as pictures, video, images, documents, etc., in a redundant multi-node
+setting. S3 is versatile enough to also be used to publish a static
+website.
+
+Garage is an opinionated object storage solutoin, we focus on the following **desirable properties**:
+
+  - **Self-contained & lightweight**: works everywhere and integrates well in existing environments to target [hyperconverged infrastructures](https://en.wikipedia.org/wiki/Hyper-converged_infrastructure).
+  - **Highly resilient**: highly resilient to network failures, network latency, disk failures, sysadmin failures.
+  - **Simple**: simple to understand, simple to operate, simple to debug.
+  - **Internet enabled**: made for multi-sites (eg. datacenters, offices, households, etc.) interconnected through regular Internet connections.
+
+We also noted that the pursuit of some other goals are detrimental to our initial goals.
+The following has been identified as **non-goals** (if these points matter to you, you should not use Garage):
+
+  - **Extreme performances**: high performances constrain a lot the design and the infrastructure; we seek performances through minimalism only.
+  - **Feature extensiveness**: we do not plan to add additional features compared to the ones provided by the S3 API.
+  - **Storage optimizations**: erasure coding or any other coding technique both increase the difficulty of placing data and synchronizing; we limit ourselves to duplication.
+  - **POSIX/Filesystem compatibility**: we do not aim at being POSIX compatible or to emulate any kind of filesystem. Indeed, in a distributed environment, such synchronizations are translated in network messages that impose severe constraints on the deployment.
+
+## Use-cases
+
+*Are you also using Garage in your organization? [Open a PR](https://git.deuxfleurs.fr/Deuxfleurs/garage) to add your use case here!*
+
+### Deuxfleurs
+
+[Deuxfleurs](https://deuxfleurs.fr) is an experimental non-profit hosting
+organization that develops Garage. Deuxfleurs is focused on building highly
+available infrastructure through redundancy in multiple geographical
+locations. They use Garage themselves for the following tasks:
+
+- Hosting of [main website](https://deuxfleurs.fr), [this website](https://garagehq.deuxfleurs.fr), as well as the personal website of many of the members of the organization
+
+- As a [Matrix media backend](https://github.com/matrix-org/synapse-s3-storage-provider)
+
+- To store personal data and shared documents through [Bagage](https://git.deuxfleurs.fr/Deuxfleurs/bagage), a homegrown WebDav-to-S3 proxy
+
+- In the Drone continuous integration platform to store task logs
+
+- As a Nix binary cache
+
+- As a backup target using `rclone`
+
+The Deuxfleurs Garage cluster is a multi-site cluster currently composed of
+4 nodes in 2 physical locations. In the future it will be expanded to at
+least 3 physical locations to fully exploit Garage's potential for high
+availability.

doc/book/src/design/index.md

@@ -1,5 +1,26 @@
 # Design
 
-The design section helps you to see Garage from a "big picture" perspective.
-It will allow you to understand if Garage is a good fit for you,
-how to better use it, how to contribute to it, what can Garage could and could not do, etc.
+The design section helps you to see Garage from a "big picture"
+perspective.  It will allow you to understand if Garage is a good fit for
+you, how to better use it, how to contribute to it, what can Garage could
+and could not do, etc.
+
+- **[Goals and use cases](goals.md):** This page explains why Garage was concieved and what practical use cases it targets.
+
+- **[Related work](related_work.md):** This pages presents the theoretical background on which Garage is built, and describes other software storage solutions and why they didn't work for us.
+
+- **[Internals](internals.md):** This page enters into more details on how Garage manages data internally.
+
+## Talks
+
+We love to talk and hear about Garage, that's why we keep a log here:
+
+  - [(fr, 2021-11-13, video) Garage : Mille et une façons de stocker vos données](https://video.tedomum.net/w/moYKcv198dyMrT8hCS5jz9) and [slides (html)](https://rfid.deuxfleurs.fr/presentations/2021-11-13/garage/) - during [RFID#1](https://rfid.deuxfleurs.fr/programme/2021-11-13/) event
+
+  - [(en, 2021-04-28) Distributed object storage is centralised](https://git.deuxfleurs.fr/Deuxfleurs/garage/raw/commit/b1f60579a13d3c5eba7f74b1775c84639ea9b51a/doc/talks/2021-04-28_spirals-team/talk.pdf)
+
+  - [(fr, 2020-12-02) Garage : jouer dans la cour des grands quand on est un hébergeur associatif](https://git.deuxfleurs.fr/Deuxfleurs/garage/raw/commit/b1f60579a13d3c5eba7f74b1775c84639ea9b51a/doc/talks/2020-12-02_wide-team/talk.pdf)
+
+*Did you write or talk about Garage? [Open a pull request](https://git.deuxfleurs.fr/Deuxfleurs/garage/) to add a link here!*
+
+

doc/book/src/design/internals.md

@@ -1,158 +1,98 @@
-**WARNING: this documentation is more a "design draft", which was written before Garage's actual implementation. The general principle is similar but details have not yet been updated.**
+# Internals
 
-#### Modules
+## Overview
 
-- `membership/`: configuration, membership management (gossip of node's presence and status), ring generation --> what about Serf (used by Consul/Nomad) : https://www.serf.io/? Seems a huge library with many features so maybe overkill/hard to integrate
-- `metadata/`: metadata management
-- `blocks/`: block management, writing, GC and rebalancing
-- `internal/`: server to server communication (HTTP server and client that reuses connections, TLS if we want, etc)
-- `api/`: S3 API
-- `web/`: web management interface
+TODO: write this section
 
-#### Metadata tables
+- The Dynamo ring (see [this paper](https://dl.acm.org/doi/abs/10.1145/1323293.1294281) and [that paper](https://www.usenix.org/conference/nsdi16/technical-sessions/presentation/eisenbud))
 
-**Objects:**
+- CRDTs (see [this paper](https://link.springer.com/chapter/10.1007/978-3-642-24550-3_29))
 
-- *Hash key:* Bucket name (string)
-- *Sort key:* Object key (string)
-- *Sort key:* Version timestamp (int)
-- *Sort key:* Version UUID (string)
-- Complete: bool
-- Inline: bool, true for objects < threshold (say 1024)
-- Object size (int)
-- Mime type (string)
-- Data for inlined objects (blob)
-- Hash of first block otherwise (string)
+- Consistency model of Garage tables
 
-*Having only a hash key on the bucket name will lead to storing all file entries of this table for a specific bucket on a single node. At the same time, it is the only way I see to rapidly being able to list all bucket entries...*
+In the meantime, you can find some information at the following links:
 
-**Blocks:**
+- [this presentation (in French)](https://git.deuxfleurs.fr/Deuxfleurs/garage/src/branch/main/doc/talks/2020-12-02_wide-team/talk.pdf)
 
-- *Hash key:* Version UUID (string)
-- *Sort key:* Offset of block in total file (int)
-- Hash of data block (string)
-
-A version is defined by the existence of at least one entry in the blocks table for a certain version UUID.
-We must keep the following invariant: if a version exists in the blocks table, it has to be referenced in the objects table.
-We explicitly manage concurrent versions of an object: the version timestamp and version UUID columns are index columns, thus we may have several concurrent versions of an object.
-Important: before deleting an older version from the objects table, we must make sure that we did a successfull delete of the blocks of that version from the blocks table.
-
-Thus, the workflow for reading an object is as follows:
-
-1. Check permissions (LDAP)
-2. Read entry in object table. If data is inline, we have its data, stop here.
-   -> if several versions, take newest one and launch deletion of old ones in background
-3. Read first block from cluster. If size <= 1 block, stop here.
-4. Simultaneously with previous step, if size > 1 block: query the Blocks table for the IDs of the next blocks
-5. Read subsequent blocks from cluster
-
-Workflow for PUT:
-
-1. Check write permission (LDAP)
-2. Select a new version UUID
-3. Write a preliminary entry for the new version in the objects table with complete = false
-4. Send blocks to cluster and write entries in the blocks table
-5. Update the version with complete = true and all of the accurate information (size, etc)
-6. Return success to the user
-7. Launch a background job to check and delete older versions
-
-Workflow for DELETE:
-
-1. Check write permission (LDAP)
-2. Get current version (or versions) in object table
-3. Do the deletion of those versions NOT IN A BACKGROUND JOB THIS TIME
-4. Return succes to the user if we were able to delete blocks from the blocks table and entries from the object table
-
-To delete a version:
-
-1. List the blocks from Cassandra
-2. For each block, delete it from cluster. Don't care if some deletions fail, we can do GC.
-3. Delete all of the blocks from the blocks table
-4. Finally, delete the version from the objects table
-
-Known issue: if someone is reading from a version that we want to delete and the object is big, the read might be interrupted. I think it is ok to leave it like this, we just cut the connection if data disappears during a read.
-
-("Soit P un problème, on s'en fout est une solution à ce problème")
-
-#### Block storage on disk
-
-**Blocks themselves:**
-
-- file path = /blobs/(first 3 hex digits of hash)/(rest of hash)
-
-**Reverse index for GC & other block-level metadata:**
-
-- file path = /meta/(first 3 hex digits of hash)/(rest of hash)
-- map block hash -> set of version UUIDs where it is referenced
-
-Usefull metadata:
-
-- list of versions that reference this block in the Casandra table, so that we can do GC by checking in Cassandra that the lines still exist
-- list of other nodes that we know have acknowledged a write of this block, usefull in the rebalancing algorithm
-
-Write strategy: have a single thread that does all write IO so that it is serialized (or have several threads that manage independent parts of the hash space). When writing a blob, write it to a temporary file, close, then rename so that a concurrent read gets a consistent result (either not found or found with whole content).
-
-Read strategy: the only read operation is get(hash) that returns either the data or not found (can do a corruption check as well and return corrupted state if it is the case). Can be done concurrently with writes.
-
-**Internal API:**
-
-- get(block hash) -> ok+data/not found/corrupted
-- put(block hash & data, version uuid + offset) -> ok/error
-- put with no data(block hash, version uuid + offset) -> ok/not found plz send data/error
-- delete(block hash, version uuid + offset) -> ok/error
-
-GC: when last ref is deleted, delete block.
-Long GC procedure: check in Cassandra that version UUIDs still exist and references this block.
-
-Rebalancing: takes as argument the list of newly added nodes.
-
-- List all blocks that we have. For each block:
-- If it hits a newly introduced node, send it to them.
-  Use put with no data first to check if it has to be sent to them already or not.
-  Use a random listing order to avoid race conditions (they do no harm but we might have two nodes sending the same thing at the same time thus wasting time).
-- If it doesn't hit us anymore, delete it and its reference list.
-
-Only one balancing can be running at a same time. It can be restarted at the beginning with new parameters.
-
-#### Membership management
-
-Two sets of nodes:
-
-- set of nodes from which a ping was recently received, with status: number of stored blocks, request counters, error counters, GC%, rebalancing%
-  (eviction from this set after say 30 seconds without ping)
-- set of nodes that are part of the system, explicitly modified by the operator using the web UI (persisted to disk),
-  is a CRDT using a version number for the value of the whole set
-
-Thus, three states for nodes:
-
-- healthy: in both sets
-- missing: not pingable but part of desired cluster
-- unused/draining: currently present but not part of the desired cluster, empty = if contains nothing, draining = if still contains some blocks
-
-Membership messages between nodes:
-
-- ping with current state + hash of current membership info -> reply with same info
-- send&get back membership info (the ids of nodes that are in the two sets): used when no local membership change in a long time and membership info hash discrepancy detected with first message (passive membership fixing with full CRDT gossip)
-- inform of newly pingable node(s) -> no result, when receive new info repeat to all (reliable broadcast)
-- inform of operator membership change -> no result, when receive new info repeat to all (reliable broadcast)
-
-Ring: generated from the desired set of nodes, however when doing read/writes on the ring, skip nodes that are known to be not pingable.
-The tokens are generated in a deterministic fashion from node IDs (hash of node id + token number from 1 to K).
-Number K of tokens per node: decided by the operator & stored in the operator's list of nodes CRDT. Default value proposal: with node status information also broadcast disk total size and free space, and propose a default number of tokens equal to 80%Free space / 10Gb. (this is all user interface)
+- [an old design draft](/working_documents/design_draft.md)
 
 
-#### Constants
+## Garbage collection
 
-- Block size: around 1MB ? --> Exoscale use 16MB chunks
-- Number of tokens in the hash ring: one every 10Gb of allocated storage
-- Threshold for storing data directly in Cassandra objects table: 1kb bytes (maybe up to 4kb?)
-- Ping timeout (time after which a node is registered as unresponsive/missing): 30 seconds
-- Ping interval: 10 seconds
-- ??
+A faulty garbage collection procedure has been the cause of
+[critical bug #39](https://git.deuxfleurs.fr/Deuxfleurs/garage/issues/39).
+This precise bug was fixed in the code, however there are potentially more
+general issues with the garbage collector being too eager and deleting things
+too early. This has been the subject of
+[PR #135](https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/135).
+This section summarizes the discussions on this topic.
 
-#### Links
+Rationale: we want to ensure Garage's safety by making sure things don't get
+deleted from disk if they are still needed. Two aspects are involved in this.
+
+### 1. Garbage collection of table entries (in `meta/` directory)
+
+The `Entry` trait used for table entries (defined in `tables/schema.rs`)
+defines a function `is_tombstone()` that returns `true` if that entry
+represents an entry that is deleted in the table. CRDT semantics by default
+keep all tombstones, because they are necessary for reconciliation: if node A
+has a tombstone that supersedes a value `x`, and node B has value `x`, A has to
+keep the tombstone in memory so that the value `x` can be properly deleted at
+node `B`. Otherwise, due to the CRDT reconciliation rule, the value `x` from B
+would flow back to A and a deleted item would reappear in the system.
+
+Here, we have some control on the nodes involved in storing Garage data.
+Therefore we have a garbage collector that is able to delete tombstones UNDER
+CERTAIN CONDITIONS. This garbage collector is implemented in `table/gc.rs`. To
+delete a tombstone, the following condition has to be met:
+
+- All nodes responsible for storing this entry are aware of the existence of
+  the tombstone, i.e. they cannot hold another version of the entry that is
+  superseeded by the tombstone. This ensures that deleting the tombstone is
+  safe and that no deleted value will come back in the system.
+
+Garage makes use of Sled's atomic operations (such as compare-and-swap and
+transactions) to ensure that only tombstones that have been correctly
+propagated to other nodes are ever deleted from the local entry tree.
+
+This GC is safe in the following sense: no non-tombstone data is ever deleted
+from Garage tables.
+
+**However**, there is an issue with the way this interacts with data
+rebalancing in the case when a partition is moving between nodes. If a node has
+some data of a partition for which it is not responsible, it has to offload it.
+However that offload process takes some time. In that interval, the GC does not
+check with that node if it has the tombstone before deleting the tombstone, so
+perhaps it doesn't have it and when the offload finally happens, old data comes
+back in the system.
+
+**PR 135 mostly fixes this** by implementing a 24-hour delay before anything is
+garbage collected in a table. This works under the assumption that rebalances
+that follow data shuffling terminate in less than 24 hours.
+
+**However**, in distributed systems, it is generally considered a bad practice
+to make assumptions that information propagates in a certain time interval:
+this consists in making a synchrony assumption, meaning that we are basically
+assuming a computing model that has much stronger properties than otherwise. To
+maximize the applicability of Garage, we would like to remove this assumption,
+and implement a system where time does not play a role. To do this, we would
+need to find a way to safely disable the GC when data is being shuffled around,
+and safely detect that the shuffling has terminated and thus the GC can be
+resumed. This introduces some complexity to the protocol and hasn't been
+tackled yet.
+
+### 2. Garbage collection of data blocks (in `data/` directory)
+
+Blocks in the data directory are reference-counted. In Garage versions before
+PR #135, blocks could get deleted from local disk as soon as their reference
+counter reached zero. We had a mechanism to not trigger this immediately at the
+rc-reaches-zero event, but the cleanup could be triggered by other means (for
+example by a block repair operation...). PR #135 added a safety measure so that
+blocks never get deleted in a 10 minute interval following the time when the RC
+reaches zero. This is a measure to make impossible race conditions such as #39.
+We would have liked to use a larger delay (e.g. 24 hours), but in the case of a
+rebalance of data, this would have led to the disk utilization to explode
+during the rebalancing, only to shrink again after 24 hours. The 10-minute
+delay is a compromise that gives good security while not having this problem of
+disk space explosion on rebalance.
 
-- CDC: <https://www.usenix.org/system/files/conference/atc16/atc16-paper-xia.pdf>
-- Erasure coding: <http://web.eecs.utk.edu/~jplank/plank/papers/CS-08-627.html>
-- [Openstack Storage Concepts](https://docs.openstack.org/arch-design/design-storage/design-storage-concepts.html)
-- [RADOS](https://ceph.com/wp-content/uploads/2016/08/weil-rados-pdsw07.pdf)

doc/book/src/design/related_work.md

@@ -1,4 +1,4 @@
-# Related Work
+# Related work
 
 ## Context
 
@@ -41,14 +41,35 @@ There were many attempts in research too. I am only thinking to [LBFS](https://p
 
 ## Existing software
 
-**[Pithos](https://github.com/exoscale/pithos) :** 
+**[MinIO](https://min.io/):** MinIO shares our *Self-contained & lightweight* goal but selected two of our non-goals: *Storage optimizations* through erasure coding and *POSIX/Filesystem compatibility* through strong consistency.
+However, by pursuing these two non-goals, MinIO do not reach our desirable properties.
+Firstly, it fails on the *Simple* property: due to the erasure coding, MinIO has severe limitations on how drives can be added or deleted from a cluster.
+Secondly, it fails on the *Internet enabled* property: due to its strong consistency, MinIO is latency sensitive.
+Furthermore, MinIO has no knowledge of "sites" and thus can not distribute data to minimize the failure of a given site.
+
+**[Openstack Swift](https://docs.openstack.org/swift/latest/):**
+OpenStack Swift at least fails on the *Self-contained & lightweight* goal.
+Starting it requires around 8GB of RAM, which is too much especially in an hyperconverged infrastructure.
+We also do not classify Swift as *Simple*.
+
+**[Ceph](https://ceph.io/ceph-storage/object-storage/):**
+This review holds for the whole Ceph stack, including the RADOS paper, Ceph Object Storage module, the RADOS Gateway, etc.
+At its core, Ceph has been designed to provide *POSIX/Filesystem compatibility* which requires strong consistency, which in turn
+makes Ceph latency-sensitive and fails our *Internet enabled* goal.
+Due to its industry oriented design, Ceph is also far from being *Simple* to operate and from being *Self-contained & lightweight* which makes it hard to integrate it in an hyperconverged infrastructure.
+In a certain way, Ceph and MinIO are closer together than they are from Garage or OpenStack Swift.
+
+**[Pithos](https://github.com/exoscale/pithos):**
 Pithos has been abandonned and should probably not used yet, in the following we explain why we did not pick their design.
 Pithos was relying as a S3 proxy in front of Cassandra (and was working with Scylla DB too).
 From its designers' mouth, storing data in Cassandra has shown its limitations justifying the project abandonment.
 They built a closed-source version 2 that does not store blobs in the database (only metadata) but did not communicate further on it.
 We considered there v2's design but concluded that it does not fit both our *Self-contained & lightweight* and *Simple* properties. It makes the development, the deployment and the operations more complicated while reducing the flexibility.
 
-**[IPFS](https://ipfs.io/) :**
+**[Riak CS](https://docs.riak.com/riak/cs/2.1.1/index.html):**
+*Not written yet*
+
+**[IPFS](https://ipfs.io/):**
 *Not written yet*
 
 ## Specific research papers

doc/book/src/development/devenv.md

@@ -1,17 +1,145 @@
 # Setup your development environment
 
-We propose the following quickstart to setup a full dev. environment as quickly as possible:
+Depending on your tastes, you can bootstrap your development environment in a traditional Rust way or through Nix.
 
-  1. Setup a rust/cargo environment. eg. `dnf install rust cargo`
-  2. Install awscli v2 by following the guide [here](https://docs.aws.amazon.com/cli/latest/userguide/install-cliv2.html).
-  3. Run `cargo build` to build the project
-  4. Run `./script/dev-cluster.sh` to launch a test cluster (feel free to read the script)
-  5. Run `./script/dev-configure.sh` to configure your test cluster with default values (same datacenter, 100 tokens)
-  6. Run `./script/dev-bucket.sh` to create a bucket named `eprouvette` and an API key that will be stored in `/tmp/garage.s3`
-  7. Run `source ./script/dev-env-aws.sh` to configure your CLI environment
-  8. You can use `garage` to manage the cluster. Try `garage --help`.
-  9. You can use the `awsgrg` alias to add, remove, and delete files. Try `awsgrg help`, `awsgrg cp /proc/cpuinfo s3://eprouvette/cpuinfo.txt`, or `awsgrg ls s3://eprouvette`. `awsgrg` is a wrapper on the `aws s3` command pre-configured with the previously generated API key (the one in `/tmp/garage.s3`) and localhost as the endpoint.
+## The Nix way
 
-Now you should be ready to start hacking on garage!
+Nix is a generic package manager we use to precisely define our development environment.
+Instructions on how to install it are given on their [Download page](https://nixos.org/download.html).
 
+Check that your installation is working by running the following commands:
 
+```
+nix-shell --version
+nix-build --version
+nix-env   --version
+```
+
+Now, you can clone our git repository (run `nix-env -iA git` if you do not have git yet):
+
+```bash
+git clone https://git.deuxfleurs.fr/Deuxfleurs/garage
+cd garage
+```
+
+*Optionnaly, you can use our nix.conf file to speed up compilations:*
+
+```bash
+sudo mkdir -p /etc/nix
+sudo cp nix/nix.conf /etc/nix/nix.conf
+sudo killall nix-daemon
+```
+
+Now you can enter our nix-shell, all the required packages will be downloaded but they will not pollute your environment outside of the shell:
+
+```bash
+nix-shell
+```
+
+You can use the traditionnal Rust development workflow:
+
+```bash
+cargo build  # compile the project
+cargo run    # execute the project
+cargo test   # run the tests
+cargo fmt    # format the project, run it before any commit!
+cargo clippy # run the linter, run it before any commit!
+```
+
+You can build the project with Nix by running:
+
+```bash
+nix-build
+```
+
+You can parallelize the build (if you use our nix.conf file, it is already automatically done).
+To use all your cores when building a derivation use `-j`, and to build multiple derivations at once use `--max-jobs`.
+The special value `auto` will be replaced by the number of cores of your computer.
+An example:
+
+```bash
+nix-build -j $(nproc) --max-jobs auto
+```
+
+Our build has multiple parameters you might want to set:
+  - `release` build with release optimisations instead of debug
+  - `target allows` for cross compilation
+  - `compileMode` can be set to test or bench to build a unit test runner
+  - `git_version` to inject the hash to display when running `garage stats`
+
+An example:
+
+```bash
+nix-build \
+  --arg release true \
+  --argstr target x86_64-unknown-linux-musl \
+  --argstr compileMode build \
+  --git_version $(git rev-parse HEAD)
+```
+
+*The result is located in `result/bin`. You can pass arguments to cross compile: check `.drone.yml` for examples.*
+
+If you modify a `Cargo.toml` or regenerate any `Cargo.lock`, you must run `cargo2nix`:
+
+```
+cargo2nix -f
+```
+
+Many tools like rclone, `mc` (minio-client), or `aws` (awscliv2) will be available in your environment and will be useful to test Garage.
+
+**This is the recommended method.**
+
+## The Rust way
+
+You need a Rust distribution installed on your computer.
+The most simple way is to install it from [rustup](https://rustup.rs).
+Please avoid using your package manager to install Rust as some tools might be outdated or missing.
+
+Now, check your Rust distribution works by running the following commands:
+
+```bash
+rustc --version
+cargo --version
+rustfmt --version
+clippy-driver --version
+```
+
+Now, you need to clone our git repository ([how to install git](https://git-scm.com/downloads)):
+
+```bash
+git clone https://git.deuxfleurs.fr/Deuxfleurs/garage
+cd garage
+```
+
+You can now use the following commands:
+
+```bash
+cargo build  # compile the project
+cargo run    # execute the project
+cargo test   # run the tests
+cargo fmt    # format the project, run it before any commit!
+cargo clippy # run the linter, run it before any commit!
+```
+
+This is specific to our project, but you will need one last tool, `cargo2nix`.
+To install it, run:
+
+```bash
+cargo install --git https://github.com/superboum/cargo2nix --branch main cargo2nix
+```
+
+You must use it every time you modify a `Cargo.toml` or regenerate a `Cargo.lock` file as follow:
+
+```bash
+cargo build  # Rebuild Cargo.lock if needed
+cargo2nix -f
+```
+
+It will output a `Cargo.nix` file which is a specific `Cargo.lock` file dedicated to Nix that is required by our CI
+which means you must include it in your commits.
+
+Later, to use our scripts and integration tests, you might need additional tools.
+These tools are listed at the end of the `shell.nix` package in the `nativeBuildInputs` part.
+It is up to you to find a way to install the ones you need on your computer.
+
+**A global drawback of this method is that it is up to you to adapt your environment to the one defined in the Nix files.**

doc/book/src/development/index.md

@@ -2,3 +2,13 @@
 
 Now that you are a Garage expert, you want to enhance it, you are in the right place!
 We discuss here how to hack on Garage, how we manage its development, etc.
+
+## Rust API (docs.rs)
+If you encounter a specific bug in Garage or plan to patch it, you may jump directly to the source code's documentation!
+
+  - [garage\_api](https://docs.rs/garage_api/latest/garage_api/) - contains the S3 standard API endpoint
+  - [garage\_model](https://docs.rs/garage_model/latest/garage_model/) - contains Garage's model built on the table abstraction
+  - [garage\_rpc](https://docs.rs/garage_rpc/latest/garage_rpc/) - contains Garage's federation protocol
+  - [garage\_table](https://docs.rs/garage_table/latest/garage_table/) - contains core Garage's CRDT datatypes
+  - [garage\_util](https://docs.rs/garage_util/latest/garage_util/) - contains garage helpers
+  - [garage\_web](https://docs.rs/garage_web/latest/garage_web/) - contains the S3 website endpoint

doc/book/src/development/miscellaneous_notes.md

@@ -0,0 +1,98 @@
+# Miscellaneous Notes
+
+## Quirks about cargo2nix/rust in Nix
+
+If you use submodules in your crate (like `crdt` and `replication` in `garage_table`), you must list them in `default.nix`
+
+The Windows target does not work. it might be solvable through [overrides](https://github.com/cargo2nix/cargo2nix/blob/master/overlay/overrides.nix). Indeed, we pass `x86_64-pc-windows-gnu` but mingw need `x86_64-w64-mingw32`
+
+We have a simple [PR on cargo2nix](https://github.com/cargo2nix/cargo2nix/pull/201) that fixes critical bugs but the project does not seem very active currently. We must use [my patched version of cargo2nix](https://github.com/superboum/cargo2nix) to enable i686 and armv6l compilation. We might need to contribute to cargo2nix in the future.
+
+
+## Nix
+
+Nix has no armv7 + musl toolchains but armv7l is backward compatible with armv6l.
+
+```bash
+cat > $HOME/.awsrc <<EOF
+export AWS_ACCESS_KEY_ID="xxx"
+export AWS_SECRET_ACCESS_KEY="xxx"
+EOF
+
+# source each time you want to send on the cache
+source ~/.awsrc
+
+# copy garage build dependencies (and not only the output)
+nix-build
+nix-store -qR --include-outputs $(nix-instantiate default.nix) 
+  | xargs nix copy --to 's3://nix?endpoint=garage.deuxfleurs.fr&region=garage'
+  
+# copy shell dependencies
+nix-build shell.nix -A inputDerivation
+nix copy $(nix-store -qR result/) --to 's3://nix?endpoint=garage.deuxfleurs.fr&region=garage' 
+```
+
+More example of nix-copy
+
+```
+# nix-build produces a result/ symlink
+nix copy result/ --to 's3://nix?endpoint=garage.deuxfleurs.fr&region=garage'
+
+# alternative ways to use nix copy
+nix copy nixpkgs.garage --to ...
+nix copy /nix/store/3rbb9qsc2w6xl5xccz5ncfhy33nzv3dp-crate-garage-0.3.0 --to ...
+```
+    
+
+Clear the cache:
+
+```bash
+mc rm --recursive --force garage/nix/
+```
+
+---
+
+A desirable `nix.conf` for a consumer:
+
+```toml
+substituters = https://cache.nixos.org https://nix.web.deuxfleurs.fr
+trusted-public-keys = cache.nixos.org-1:6NCHdD59X431o0gWypbMrAURkbJ16ZPMQFGspcDShjY= nix.web.deuxfleurs.fr:eTGL6kvaQn6cDR/F9lDYUIP9nCVR/kkshYfLDJf1yKs=
+```
+
+And now, whenever you run a command like:
+
+```
+nix-shell
+nix-build
+```
+
+Our cache will be checked.
+
+### Some references about Nix
+
+
+ - https://doc.rust-lang.org/nightly/rustc/platform-support.html
+ - https://nix.dev/tutorials/cross-compilation
+ - https://nixos.org/manual/nix/unstable/package-management/s3-substituter.html
+ - https://fzakaria.com/2020/09/28/nix-copy-closure-your-nix-shell.html
+ - http://www.lpenz.org/articles/nixchannel/index.html
+
+
+## Drone
+
+Do not try to set a build as trusted from the interface or the CLI tool,
+your request would be ignored. Instead, directly edit the database (table `repos`, column `repo_trusted`).
+
+Drone can do parallelism both at the step and the pipeline level. At the step level, parallelism is restricted to the same runner.
+
+## Building Docker containers
+
+We were:
+  - Unable to use the official Docker plugin because
+    - it requires to mount docker socket in the container but it is not recommended
+    - you cant set the platform when building
+  - Unable to use buildah because it needs `CLONE_USERNS` capability
+  - Unable to use the kaniko plugin for Drone as we can't set the target platform
+  - Unable to use the kaniko container provided by Google as we can't run arbitrary logic: we need to put our secret in .docker/config.json.
+
+Finally we chose to build kaniko through nix and use it in a `nix-shell`.

doc/book/src/development/release_process.md

@@ -0,0 +1,195 @@
+# Release process
+
+Before releasing a new version of Garage, our code pass through a succession of checks and transformations.
+We define them as our release process.
+
+## Trigger and classify a release
+
+While we run some tests on every commits, we do not make a release for all of them.
+
+A release can be triggered manually by "promoting" a successful build.
+Otherwise, every weeks, a release build is triggered on the `main` branch.
+
+If the build is from a tag following the regex: `v[0-9]+\.[0-9]+\.[0-9]+`, it will be listed as stable.
+If it is a tag but with a different format, it will be listed as Extra.
+Otherwise, if it is a commit, it will be listed as development.
+This logic is defined in `nix/build_index.nix`.
+
+## Testing
+
+For each commit, we first pass the code to a formatter (rustfmt) and a linter (clippy).
+Then we try to build it in debug mode and run both unit tests and our integration tests.
+
+Additionnaly, when releasing, our integration tests are run on the release build for amd64 and i686.
+
+## Generated Artifacts
+
+We generate the following binary artifacts for now:
+  - **architecture**: amd64, i686, aarch64, armv6
+  - **os**: linux
+  - **format**: static binary, docker container
+
+Additionnaly we also build two web pages:
+  - the documentation (this website)
+  - [the release page](https://garagehq.deuxfleurs.fr/releases.html)
+
+We publish the static binaries on our own garage cluster (you can access them through the releases page)
+and the docker containers on Docker Hub.
+
+## Automation
+
+We automated our release process with Nix and Drone to make it more reliable.
+Here we describe how we have done in case you want to debug or improve it.
+
+### Caching build steps
+
+To speed up the CI, we use the caching feature provided by Nix.
+
+You can benefit from it by using our provided `nix.conf` as recommended or by simply adding the following lines to your file:
+
+```toml
+substituters = https://cache.nixos.org https://nix.web.deuxfleurs.fr
+trusted-public-keys = cache.nixos.org-1:6NCHdD59X431o0gWypbMrAURkbJ16ZPMQFGspcDShjY= nix.web.deuxfleurs.fr:eTGL6kvaQn6cDR/F9lDYUIP9nCVR/kkshYfLDJf1yKs=
+```
+
+Sending to the cache is done through `nix copy`, for example:
+
+```bash
+nix copy --to 's3://nix?endpoint=garage.deuxfleurs.fr&region=garage&secret-key=/etc/nix/signing-key.sec' result
+```
+
+*Note that you need the signing key. In our case, it is stored as a secret in Drone.*
+
+The previous command will only send the built packet and not its dependencies.
+To send its dependency, a tool named `nix-copy-closure` has been created but it is not compatible with the S3 protocol.
+
+Instead, you can use the following commands to list all the runtime dependencies:
+
+```bash
+nix copy \
+  --to 's3://nix?endpoint=garage.deuxfleurs.fr&region=garage&secret-key=/etc/nix/signing-key.sec' \
+  $(nix-store -qR result/)
+```
+
+*We could also write this expression with xargs but this tool is not available in our container.*
+
+But in certain cases, we want to cache compile time dependencies also.
+For example, the Nix project does not provide binaries for cross compiling to i686 and thus we need to compile gcc on our own.
+We do not want to compile gcc each time, so even if it is a compile time dependency, we want to cache it.
+
+This time, the command is a bit more involved:
+
+```bash
+nix copy --to \
+  's3://nix?endpoint=garage.deuxfleurs.fr&region=garage&secret-key=/etc/nix/signing-key.sec' \
+   $(nix-store -qR --include-outputs \
+     $(nix-instantiate))
+```
+
+This is the command we use in our CI as we expect the final binary to change, so we mainly focus on
+caching our development dependencies.
+
+*Currently there is no automatic garbage collection of the cache: we should monitor its growth.
+Hopefully, we can erase it totally without breaking any build, the next build will only be slower.*
+
+In practise, we concluded that we do not want to cache all the compilation dependencies.
+Instead, we want to cache the toolchain we use to build Garage each time we change it.
+So we removed from Drone any automatic update of the cache and instead handle them manually with:
+
+```
+source ~/.awsrc
+nix-shell --run 'refresh_toolchain'
+```
+
+Internally, it will run `nix-build` on  `nix/toolchain.nix` and send the output plus its depedencies to the cache.
+
+To erase the cache:
+
+```
+mc rm --recursive --force 'garage/nix/'
+```
+
+### Publishing Garage
+
+We defined our publishing logic in Nix, mostly as shell hooks.
+You can inspect them in `shell.nix` to see exactly how.
+Here, we will give a quick explanation on how to use them to manually publish a release.
+
+Supposing you just have built garage as follow:
+
+```bash
+nix-build --arg release true
+```
+
+To publish a static binary in `result/bin` on garagehq, run:
+
+```bash
+export AWS_ACCESS_KEY_ID=xxx
+export AWS_SECRET_ACCESS_KEY=xxx
+export DRONE_TAG=handcrafted-1.0.0 # or DRONE_COMMIT
+export TARGET=x86_64-unknown-linux-musl
+
+nix-shell --run to_s3
+```
+
+To create and publish a docker container, run:
+
+```bash
+export DOCKER_AUTH='{ "auths": { "https://index.docker.io/v1/": { "auth": "xxxx" }}}'
+export DOCKER_PLATFORM='linux/amd64' # check GOARCH and GOOS from golang.org
+export CONTAINER_NAME='me/amd64_garage'
+export CONTAINER_TAG='handcrafted-1.0.0'
+
+nix-shell --run to_docker
+```
+
+To rebuild the release page, run:
+```bash
+export AWS_ACCESS_KEY_ID=xxx
+export AWS_SECRET_ACCESS_KEY=xxx
+
+nix-shell --run refresh_index
+```
+
+If you want to compile for different architectures, you will need to repeat all these commands for each architecture.
+
+**In practise, and except for debugging, you will never directly run these commands. Release is handled by drone**
+
+### Drone
+
+Our instance is available at [https://drone.deuxfleurs.fr](https://drone.deuxfleurs.fr).  
+You need an account on [https://git.deuxfleurs.fr](https://git.deuxfleurs.fr) to use it.
+
+**Drone CLI** - Drone has a CLI tool to interact with.
+It can be downloaded from its Github [release page](https://github.com/drone/drone-cli/releases).
+
+To communicate with our instance, you must setup some environment variables.
+You can get them from your [Account Settings](https://drone.deuxfleurs.fr/account).
+
+To make drone easier to use, you could create a `~/.dronerc` that you could source each time you want to use it.
+
+```
+export DRONE_SERVER=https://drone.deuxfleurs.fr
+export DRONE_TOKEN=xxx
+drone info
+```
+
+The CLI tool is very self-discoverable, just append `--help` to each subcommands.
+Start with:
+
+```bash
+drone --help
+```
+
+**.drone.yml** - The builds steps are defined in `.drone.yml`.  
+You can not edit this file without resigning it.
+
+To sign it, you must be a maintainer and then run:
+
+```bash
+drone sign --save Deuxfleurs/garage
+```
+
+Looking at the file, you will see that most of the commands are `nix-shell` and `nix-build` commands with various parameters.
+
+

doc/book/src/development/scripts.md

@@ -0,0 +1,113 @@
+# Development scripts
+
+We maintain a `script/` folder that contains some useful script to ease testing on Garage.
+
+A fully integrated script, `test-smoke.sh`, runs some basic tests on various tools such as minio client, awscli and rclone.
+To run it, enter a `nix-shell` (or install all required tools) and simply run:
+
+```bash
+nix-build # or cargo build
+./script/test-smoke.sh
+```
+
+If something fails, you can find useful logs in `/tmp/garage.log`.
+You can inspect the generated configuration and local data created by inspecting your `/tmp` directory:
+the script creates files and folder prefixed with the name "garage".
+
+## Bootstrapping a test cluster
+
+Under the hood `test-smoke.sh` uses multiple helpers scripts you can also run in case you want to manually test Garage.
+In this section, we introduce 3 scripts to quickly bootstrap a full test cluster with 3 instances.
+
+### 1. Start each daemon
+
+```bash
+./script/dev-cluster.sh
+```
+
+This script spawns 3 Garage instances with 3 configuration files.
+You can inspect the detailed configuration, including ports, by inspecting `/tmp/config.1` (change 1 by the instance number you want).
+
+This script also spawns a simple HTTPS reverse proxy through `socat` for the S3 endpoint that listens on port `4443`.
+Some libraries might require a TLS endpoint to work, refer to our issue [#64](https://git.deuxfleurs.fr/Deuxfleurs/garage/issues/64) for more detailed information on this subject.
+
+This script covers the [Launching the garage server](/quick_start/index.html#launching-the-garage-server) section of our Quick start page.
+
+### 2. Make them join the cluster
+
+```bash
+./script/dev-configure.sh
+```
+
+This script will configure each instance by assigning them a zone (`dc1`) and a weight (`1`).
+
+This script covers the [Configuring your Garage node](/quick_start/index.html#configuring-your-garage-node) section of our Quick start page.
+
+### 3. Create a key and a bucket
+
+```bash
+./script/dev-bucket.sh
+```
+
+This script will create a bucket named `eprouvette` with a key having read and write rights on this bucket.
+The key is stored in a filed named `/tmp/garage.s3` and can be used by the following tools to pre-configure them.
+
+This script covers the [Creating buckets and keys](/quick_start/index.html#creating-buckets-and-keys) section of our Quick start page.
+
+## Handlers for generic tools
+
+We provide wrappers for some CLI tools that configure themselves for your development cluster.
+They are meant to save you some configuration time as to use them, you are only required to source the right file.
+
+### awscli
+
+```bash
+source ./script/dev-env-aws.sh
+
+# some examples
+aws s3 ls s3://eprouvette
+aws s3 cp /proc/cpuinfo s3://eprouvette/cpuinfo.txt
+```
+
+### minio-client
+
+
+```bash
+source ./script/dev-env-mc.sh
+
+# some examples
+mc ls garage/
+mc cp /proc/cpuinfo garage/eprouvette/cpuinfo.txt
+```
+
+### rclone
+
+```bash
+source ./script/dev-env-rclone.sh
+
+# some examples
+rclone lsd garage:
+rclone copy /proc/cpuinfo garage:eprouvette/cpuinfo.txt
+```
+
+### s3cmd
+
+```bash
+source ./script/dev-env-s3cmd.sh
+
+# some examples
+s3cmd ls
+s3cmd put /proc/cpuinfo s3://eprouvette/cpuinfo.txt
+```
+
+### duck
+
+*Warning! Duck is not yet provided by nix-shell.*
+
+```bash
+source ./script/dev-env-duck.sh
+
+# some examples
+duck --list garage:/
+duck --upload garage:/eprouvette/ /proc/cpuinfo
+```

doc/book/src/intro.md

@@ -5,102 +5,75 @@
 </p>
 
 <p align="center" style="text-align:center;">
-	[ <a href="https://git.deuxfleurs.fr/Deuxfleurs/garage">Git repository</a>
+	[ <a href="https://garagehq.deuxfleurs.fr/_releases.html">Download</a>
+	| <a href="https://git.deuxfleurs.fr/Deuxfleurs/garage">Git repository</a>
 	| <a href="https://matrix.to/#/%23garage:deuxfleurs.fr">Matrix channel</a>
 	| <a href="https://drone.deuxfleurs.fr/Deuxfleurs/garage">Drone CI</a>
 	]
 </p>
 
-```
-This very website is hosted using Garage. In other words: the doc is the PoC!
-```
 
-# The Garage Geo-Distributed Data Store
+# Data resiliency for everyone
 
-Garage is a lightweight geo-distributed data store.
-It comes from the observation that despite numerous object stores
-many people have broken data management policies (backup/replication on a single site or none at all).
-To promote better data management policies, we focused on the following **desirable properties**:
+Garage is an **open-source** distributed **storage service** you can **self-host** to fullfill many needs:
 
-  - **Self-contained & lightweight**: works everywhere and integrates well in existing environments to target [hyperconverged infrastructures](https://en.wikipedia.org/wiki/Hyper-converged_infrastructure).
-  - **Highly resilient**: highly resilient to network failures, network latency, disk failures, sysadmin failures.
-  - **Simple**: simple to understand, simple to operate, simple to debug.
-  - **Internet enabled**: made for multi-sites (eg. datacenters, offices, households, etc.) interconnected through regular Internet connections.
+<p align="center" style="text-align:center; margin-bottom: 5rem;">
+<img alt="Summary of the possible usages with a related icon: host a website, store media and backup target" src="img/usage.svg" />
+</p>
 
-We also noted that the pursuit of some other goals are detrimental to our initial goals.
-The following has been identified as **non-goals** (if these points matter to you, you should not use Garage):
+<p align="center" style="text-align:center; margin-bottom: 5rem;">
+<a href="/design/goals.html#use-cases">⮞ learn more about use cases ⮜</a>
+</p>
 
-  - **Extreme performances**: high performances constrain a lot the design and the infrastructure; we seek performances through minimalism only.
-  - **Feature extensiveness**: complete implementation of the S3 API or any other API to make garage a drop-in replacement is not targeted as it could lead to decisions impacting our desirable properties.
-  - **Storage optimizations**: erasure coding or any other coding technique both increase the difficulty of placing data and synchronizing; we limit ourselves to duplication.
-  - **POSIX/Filesystem compatibility**: we do not aim at being POSIX compatible or to emulate any kind of filesystem. Indeed, in a distributed environment, such synchronizations are translated in network messages that impose severe constraints on the deployment.
+Garage implements the **[Amazon S3 API](https://docs.aws.amazon.com/AmazonS3/latest/API/Welcome.html)** and thus is already **compatible** with many applications:
 
-## Supported and planned protocols
+<p align="center" style="text-align:center; margin-bottom: 8rem;">
+<img alt="Garage is already compatible with Nextcloud, Mastodon, Matrix Synapse, Cyberduck, RClone and Peertube" src="img/software.svg" />
+</p>
 
-Garage speaks (or will speak) the following protocols:
+<p align="center" style="text-align:center; margin-bottom: 5rem;">
+<a href="/connect/index.html">⮞ learn more about integrations ⮜</a>
+</p>
 
-  - [S3](https://docs.aws.amazon.com/AmazonS3/latest/API/Welcome.html) - *SUPPORTED* - Enable applications to store large blobs such as pictures, video, images, documents, etc. S3 is versatile enough to also be used to publish a static website.
-  - [IMAP](https://github.com/go-pluto/pluto) - *PLANNED* - email storage is quite complex to get good performances.
-To keep performances optimal, most IMAP servers only support on-disk storage.
-We plan to add logic to Garage to make it a viable solution for email storage.
-  - *More to come*
 
-## Use Cases
+Garage provides **data resiliency** by **replicating** data 3x over **distant** servers:
 
-**[Deuxfleurs](https://deuxfleurs.fr):** Garage is used by Deuxfleurs which is a non-profit hosting organization.
-Especially, it is used to host their main website, this documentation and some of its members' blogs. 
-Additionally, Garage is used as a [backend for Nextcloud](https://docs.nextcloud.com/server/20/admin_manual/configuration_files/primary_storage.html). 
-Deuxfleurs also plans to use Garage as their [Matrix's media backend](https://github.com/matrix-org/synapse-s3-storage-provider) and as the backend of [OCIS](https://github.com/owncloud/ocis).
+<p align="center" style="text-align:center; margin-bottom: 5rem;">
+<img alt="An example deployment on a map with servers in 5 zones: UK, France, Belgium, Germany and Switzerland. Each chunk of data is replicated in 3 of these 5 zones." src="img/map.svg" />
+</p>
 
-*Are you using Garage? [Open a pull request](https://git.deuxfleurs.fr/Deuxfleurs/garage/) to add your organization here!*
+<p align="center" style="text-align:center; margin-bottom: 5rem;">
+<a href="/design/index.html">⮞ learn more about our design ⮜</a>
+</p>
 
-## Comparison to existing software
+Did you notice that *this website* is hosted and served by Garage?
 
-**[MinIO](https://min.io/):** MinIO shares our *Self-contained & lightweight* goal but selected two of our non-goals: *Storage optimizations* through erasure coding and *POSIX/Filesystem compatibility* through strong consistency.
-However, by pursuing these two non-goals, MinIO do not reach our desirable properties.
-Firstly, it fails on the *Simple* property: due to the erasure coding, MinIO has severe limitations on how drives can be added or deleted from a cluster.
-Secondly, it fails on the *Internet enabled* property: due to its strong consistency, MinIO is latency sensitive.
-Furthermore, MinIO has no knowledge of "sites" and thus can not distribute data to minimize the failure of a given site.
+## Keeping requirements low
 
-**[Openstack Swift](https://docs.openstack.org/swift/latest/):**
-OpenStack Swift at least fails on the *Self-contained & lightweight* goal.
-Starting it requires around 8GB of RAM, which is too much especially in an hyperconverged infrastructure.
-We also do not classify Swift as *Simple*.
+We worked hard to keep requirements as low as possible as we target the largest possible public.
 
-**[Ceph](https://ceph.io/ceph-storage/object-storage/):**
-This review holds for the whole Ceph stack, including the RADOS paper, Ceph Object Storage module, the RADOS Gateway, etc.
-At its core, Ceph has been designed to provide *POSIX/Filesystem compatibility* which requires strong consistency, which in turn
-makes Ceph latency-sensitive and fails our *Internet enabled* goal. 
-Due to its industry oriented design, Ceph is also far from being *Simple* to operate and from being *Self-contained & lightweight* which makes it hard to integrate it in an hyperconverged infrastructure.
-In a certain way, Ceph and MinIO are closer together than they are from Garage or OpenStack Swift.
+  * **CPU:** any x86\_64 CPU from the last 10 years, ARMv7 or ARMv8.
+  * **RAM:** 1GB
+  * **Disk Space:** at least 16GB
+  * **Network:** 200ms or less, 50 Mbps or more
+  * **Heterogeneous hardware:** build a cluster with whatever second-hand machines are available
 
-*More comparisons are available in our [Related Work](design/related_work.md) chapter.*
+*For the network, as we do not use consensus algorithms like Paxos or Raft, Garage is not as latency sensitive.*
+*Thanks to Rust and its zero-cost abstractions, we keep CPU and memory low.*
 
-## Other Resources
+## Built on the shoulder of giants
 
-This website is not the only source of information about Garage!
-We reference here other places on the Internet where you can learn more about Garage.
+  - [Dynamo: Amazon’s Highly Available Key-value Store ](https://dl.acm.org/doi/abs/10.1145/1323293.1294281) by DeCandia et al.
+  - [Conflict-Free Replicated Data Types](https://link.springer.com/chapter/10.1007/978-3-642-24550-3_29) by Shapiro et al.
+  - [Maglev: A Fast and Reliable Software Network Load Balancer](https://www.usenix.org/conference/nsdi16/technical-sessions/presentation/eisenbud) by Eisenbud et al.
 
-### Rust API (docs.rs)
+## Talks
 
-If you encounter a specific bug in Garage or plan to patch it, you may jump directly to the source code's documentation!
+  - [(fr, 2021-11-13, video) Garage : Mille et une façons de stocker vos données](https://video.tedomum.net/w/moYKcv198dyMrT8hCS5jz9) and [slides (html)](https://rfid.deuxfleurs.fr/presentations/2021-11-13/garage/) - during [RFID#1](https://rfid.deuxfleurs.fr/programme/2021-11-13/) event
 
-  - [garage\_api](https://docs.rs/garage_api/latest/garage_api/) - contains the S3 standard API endpoint
-  - [garage\_model](https://docs.rs/garage_model/latest/garage_model/) - contains Garage's model built on the table abstraction
-  - [garage\_rpc](https://docs.rs/garage_rpc/latest/garage_rpc/) - contains Garage's federation protocol
-  - [garage\_table](https://docs.rs/garage_table/latest/garage_table/) - contains core Garage's CRDT datatypes
-  - [garage\_util](https://docs.rs/garage_util/latest/garage_util/) - contains garage helpers
-  - [garage\_web](https://docs.rs/garage_web/latest/garage_web/) - contains the S3 website endpoint
+  - [(en, 2021-04-28, pdf) Distributed object storage is centralised](https://git.deuxfleurs.fr/Deuxfleurs/garage/raw/commit/b1f60579a13d3c5eba7f74b1775c84639ea9b51a/doc/talks/2021-04-28_spirals-team/talk.pdf)
 
-### Talks
-
-We love to talk and hear about Garage, that's why we keep a log here:
-
-  - [(en, 2021-04-28) Distributed object storage is centralised](https://git.deuxfleurs.fr/Deuxfleurs/garage/raw/commit/b1f60579a13d3c5eba7f74b1775c84639ea9b51a/doc/talks/2021-04-28_spirals-team/talk.pdf)
-
-  - [(fr, 2020-12-02) Garage : jouer dans la cour des grands quand on est un hébergeur associatif](https://git.deuxfleurs.fr/Deuxfleurs/garage/raw/commit/b1f60579a13d3c5eba7f74b1775c84639ea9b51a/doc/talks/2020-12-02_wide-team/talk.pdf)
-
-*Did you write or talk about Garage? [Open a pull request](https://git.deuxfleurs.fr/Deuxfleurs/garage/) to add a link here!*
+  - [(fr, 2020-12-02, pdf) Garage : jouer dans la cour des grands quand on est un hébergeur associatif](https://git.deuxfleurs.fr/Deuxfleurs/garage/raw/commit/b1f60579a13d3c5eba7f74b1775c84639ea9b51a/doc/talks/2020-12-02_wide-team/talk.pdf)
 
 ## Community
 
@@ -111,3 +84,18 @@ Our code repository and issue tracker, which is the place where you should repor
 
 Garage's source code, is released under the [AGPL v3 License](https://www.gnu.org/licenses/agpl-3.0.en.html).
 Please note that if you patch Garage and then use it to provide any service over a network, you must share your code!
+
+# Sponsors and funding
+
+The Deuxfleurs association has received a grant from [NGI POINTER](https://pointer.ngi.eu/), to fund 3 people working on Garage full-time for a year: from October 2021 to September 2022.
+
+<div style="display: flex; justify-content: space-around">
+  <a href="https://pointer.ngi.eu/">
+    <img style="height:100px" src="img/ngi-logo.png" alt="NGI Pointer logo">
+  </a>
+  <a href="https://ec.europa.eu/programmes/horizon2020/what-horizon-2020">
+    <img style="height:100px" src="img/eu-flag-logo.png" alt="EU flag logo">
+  </a>
+</div>
+
+_This project has received funding from the European Union’s Horizon 2020 research and innovation programme within the framework of the NGI-POINTER Project funded under grant agreement N° 871528._

doc/book/src/quick_start/index.md

@@ -6,31 +6,33 @@ and how to interact with it.
 
 Our goal is to introduce you to Garage's workflows.
 Following this guide is recommended before moving on to
-[configuring a real-world deployment](../cookbook/real_world.md).
+[configuring a multi-node cluster](../cookbook/real_world.md).
 
-Note that this kind of deployment should not be used in production, as it provides
-no redundancy for your data!
-We will also skip intra-cluster TLS configuration, meaning that if you add nodes
-to your cluster, communication between them will not be secure.
+Note that this kind of deployment should not be used in production,
+as it provides no redundancy for your data!
 
 ## Get a binary
 
 Download the latest Garage binary from the release pages on our repository:
 
-<https://git.deuxfleurs.fr/Deuxfleurs/garage/releases>
+<https://garagehq.deuxfleurs.fr/_releases.html>
 
 Place this binary somewhere in your `$PATH` so that you can invoke the `garage`
 command directly (for instance you can copy the binary in `/usr/local/bin`
 or in `~/.local/bin`).
 
 If a binary of the last version is not available for your architecture,
+or if you want a build customized for your system,
 you can [build Garage from source](../cookbook/from_source.md).
 
 
 ## Writing a first configuration file
 
 This first configuration file should allow you to get started easily with the simplest
-possible Garage deployment:
+possible Garage deployment.
+**Save it as `/etc/garage.toml`.**
+You can also store it somewhere else, but you will have to specify `-c path/to/garage.toml`
+at each invocation of the `garage` binary (for example: `garage -c ./garage.toml server`, `garage -c ./garage.toml status`).
 
 ```toml
 metadata_dir = "/tmp/meta"
@@ -39,22 +41,26 @@ data_dir = "/tmp/data"
 replication_mode = "none"
 
 rpc_bind_addr = "[::]:3901"
+rpc_public_addr = "127.0.0.1:3901"
+rpc_secret = "1799bccfd7411eddcf9ebd316bc1f5287ad12a68094e1c6ac6abde7e6feae1ec"
 
-bootstrap_peers = [
-	"127.0.0.1:3901",
-]
+bootstrap_peers = []
 
 [s3_api]
 s3_region = "garage"
 api_bind_addr = "[::]:3900"
+root_domain = ".s3.garage.localhost"
 
 [s3_web]
 bind_addr = "[::]:3902"
-root_domain = ".web.garage"
+root_domain = ".web.garage.localhost"
 index = "index.html"
 ```
 
-Save your configuration file as `garage.toml`.
+The `rpc_secret` value provided above is just an example. It will work, but in
+order to secure your cluster you will need to use another one. You can generate
+such a value with `openssl rand -hex 32`.
+
 
 As you can see in the `metadata_dir` and `data_dir` parameters, we are saving Garage's data
 in `/tmp` which gets erased when your system reboots. This means that data stored on this
@@ -67,15 +73,15 @@ your data to be persisted properly.
 Use the following command to launch the Garage server with our configuration file:
 
 ```
-RUST_LOG=garage=info garage server -c garage.toml
+RUST_LOG=garage=info garage server
 ```
 
 You can tune Garage's verbosity as follows (from less verbose to more verbose):
 
 ```
-RUST_LOG=garage=info garage server -c garage.toml
-RUST_LOG=garage=debug garage server -c garage.toml
-RUST_LOG=garage=trace garage server -c garage.toml
+RUST_LOG=garage=info garage server
+RUST_LOG=garage=debug garage server
+RUST_LOG=garage=trace garage server
 ```
 
 Log level `info` is recommended for most use cases.
@@ -85,11 +91,12 @@ Log level `debug` can help you check why your S3 API calls are not working.
 ## Checking that Garage runs correctly
 
 The `garage` utility is also used as a CLI tool to configure your Garage deployment.
-It tries to connect to a Garage server through the RPC protocol, by default looking
-for a Garage server at `localhost:3901`.
+It uses values from the TOML configuration file to find the Garage daemon running on the
+local node, therefore if your configuration file is not at `/etc/garage.toml` you will
+again have to specify `-c path/to/garage.toml`.
 
-Since our deployment already binds to port 3901, the following command should be sufficient
-to show Garage's status:
+If the `garage` CLI is able to correctly detect the parameters of your local Garage node,
+the following command should be enough to show the status of your cluster:
 
 ```
 garage status
@@ -98,13 +105,14 @@ garage status
 This should show something like this:
 
 ```
-Healthy nodes:
-2a638ed6c775b69a…	linuxbox	127.0.0.1:3901	UNCONFIGURED/REMOVED
+==== HEALTHY NODES ====
+ID                 Hostname  Address         Tag                   Zone  Capacity
+563e1ac825ee3323…  linuxbox  127.0.0.1:3901  NO ROLE ASSIGNED
 ```
 
-## Configuring your Garage node
+## Creating a cluster layout
 
-Configuring the nodes in a Garage deployment means informing Garage
+Creating a cluster layout for a Garage deployment means informing Garage
 of the disk space available on each node of the cluster
 as well as the zone (e.g. datacenter) each machine is located in.
 
@@ -112,14 +120,18 @@ For our test deployment, we are using only one node. The way in which we configu
 it does not matter, you can simply write:
 
 ```bash
-garage node configure -z dc1 -c 1 <node_id>
+garage layout assign -z dc1 -c 1 <node_id>
 ```
 
 where `<node_id>` corresponds to the identifier of the node shown by `garage status` (first column).
 You can enter simply a prefix of that identifier.
-For instance here you could write just `garage node configure -z dc1 -c 1 2a63`.
+For instance here you could write just `garage layout assign -z dc1 -c 1 563e`.
 
+The layout then has to be applied to the cluster, using:
 
+```bash
+garage layout apply
+```
 
 
 ## Creating buckets and keys
@@ -190,7 +202,7 @@ Now that we have a bucket and a key, we need to give permissions to the key on t
 ```
 garage bucket allow \
   --read \
-  --write 
+  --write \
   nextcloud-bucket \
   --key nextcloud-app-key
 ```
@@ -263,5 +275,5 @@ The following tools can also be used to send and recieve files from/to Garage:
 - [Cyberduck](https://cyberduck.io/)
 - [`s3cmd`](https://s3tools.org/s3cmd)
 
-Refer to the ["configuring clients"](../cookbook/clients.md) page to learn how to configure
-these clients to interact with a Garage server.
+Refer to the ["Integrations" section](../connect/index.md) to learn how to
+configure application and command line utilities to integrate with Garage.

doc/book/src/reference_manual/configuration.md

@@ -10,31 +10,29 @@ block_size = 1048576
 
 replication_mode = "3"
 
+compression_level = 1
+
+rpc_secret = "4425f5c26c5e11581d3223904324dcb5b5d5dfb14e5e7f35e38c595424f5f1e6"
 rpc_bind_addr = "[::]:3901"
+rpc_public_addr = "[fc00:1::1]:3901"
 
 bootstrap_peers = [
-  "[fc00:1::1]:3901",
-  "[fc00:1::2]:3901",
-  "[fc00:B::1]:3901",
-  "[fc00:F::1]:3901",
+    "563e1ac825ee3323aa441e72c26d1030d6d4414aeb3dd25287c531e7fc2bc95d@[fc00:1::1]:3901",
+    "86f0f26ae4afbd59aaf9cfb059eefac844951efd5b8caeec0d53f4ed6c85f332[fc00:1::2]:3901",
+    "681456ab91350f92242e80a531a3ec9392cb7c974f72640112f90a600d7921a4@[fc00:B::1]:3901",
+    "212fd62eeaca72c122b45a7f4fa0f55e012aa5e24ac384a72a3016413fa724ff@[fc00:F::1]:3901",
 ]
 
 consul_host = "consul.service"
 consul_service_name = "garage-daemon"
 
-max_concurrent_rpc_requests = 12
-
 sled_cache_capacity = 134217728
 sled_flush_every_ms = 2000
 
-[rpc_tls]
-ca_cert = "/etc/garage/pki/garage-ca.crt"
-node_cert = "/etc/garage/pki/garage.crt"
-node_key = "/etc/garage/pki/garage.key"
-
 [s3_api]
-s3_region = "garage"
 api_bind_addr = "[::]:3900"
+s3_region = "garage"
+root_domain = ".s3.garage"
 
 [s3_web]
 bind_addr = "[::]:3902"
@@ -63,10 +61,15 @@ when [configuring it](../getting_started/05_cluster.md).
 
 #### `block_size`
 
-Garage splits stored objects in consecutive chunks of size `block_size` (except the last
-one which might be standard). The default size is 1MB and should work in most cases.
-If you are interested in tuning this, feel free to do so (and remember to report your
-findings to us!)
+Garage splits stored objects in consecutive chunks of size `block_size`
+(except the last one which might be smaller). The default size is 1MB and
+should work in most cases.  If you are interested in tuning this, feel free
+to do so (and remember to report your findings to us!). If this value is
+changed for a running Garage installation, only files newly uploaded will be
+affected. Previously uploaded files will remain available. This however
+means that chunks from existing files will not be deduplicated with chunks
+from newly uploaded files, meaning you might use more storage space that is
+optimally possible.
 
 #### `replication_mode`
 
@@ -97,6 +100,38 @@ Never run a Garage cluster where that is not the case.**
 Changing the `replication_mode` of a cluster might work (make sure to shut down all nodes
 and changing it everywhere at the time), but is not officially supported.
 
+### `compression_level`
+
+Zstd compression level to use for storing blocks.
+
+Values between `1` (faster compression) and `19` (smaller file) are standard compression
+levels for zstd. From `20` to `22`, compression levels are referred as "ultra" and must be
+used with extra care as it will use lot of memory. A value of `0` will let zstd choose a
+default value (currently `3`). Finally, zstd has also compression designed to be faster
+than default compression levels, they range from `-1` (smaller file) to `-99` (faster 
+compression).
+
+If you do not specify a `compression_level` entry, garage will set it to `1` for you. With
+this parameters, zstd consumes low amount of cpu and should work faster than line speed in
+most situations, while saving some space and intra-cluster
+bandwidth.
+
+If you want to totally deactivate zstd in garage, you can pass the special value `'none'`. No
+zstd related code will be called, your chunks will be stored on disk without any processing.
+
+Compression is done synchronously, setting a value too high will add latency to write queries.
+
+This value can be different between nodes, compression is done by the node which receive the
+API call.
+
+#### `rpc_secret`
+
+Garage uses a secret key that is shared between all nodes of the cluster
+in order to identify these nodes and allow them to communicate together.
+This key should be specified here in the form of a 32-byte hex-encoded
+random string. Such a string can be generated with a command
+such as `openssl rand -hex 32`.
+
 #### `rpc_bind_addr`
 
 The address and port on which to bind for inter-cluster communcations
@@ -106,10 +141,28 @@ the node, even in the case of a NAT: the NAT should be configured to forward the
 port number to the same internal port nubmer. This means that if you have several nodes running
 behind a NAT, they should each use a different RPC port number.
 
+#### `rpc_public_addr`
+
+The address and port that other nodes need to use to contact this node for
+RPC calls.  **This parameter is optional but recommended.** In case you have
+a NAT that binds the RPC port to a port that is different on your public IP,
+this field might help making it work.
+
 #### `bootstrap_peers`
 
-A list of IPs and ports on which to contact other Garage peers of this cluster.
-This should correspond to the RPC ports set up with `rpc_bind_addr`.
+A list of peer identifiers on which to contact other Garage peers of this cluster.
+These peer identifiers have the following syntax:
+
+```
+<node public key>@<node public IP or hostname>:<port>
+```
+
+In the case where `rpc_public_addr` is correctly specified in the
+configuration file, the full identifier of a node including IP and port can
+be obtained by running `garage node id` and then included directly in the
+`bootstrap_peers` list of other nodes.  Otherwise, only the node's public
+key will be returned by `garage node id` and you will have to add the IP
+yourself.
 
 #### `consul_host` and `consul_service_name`
 
@@ -121,12 +174,6 @@ The `consul_host` parameter should be set to the hostname of the Consul server,
 and `consul_service_name` should be set to the service name under which Garage's
 RPC ports are announced.
 
-#### `max_concurrent_rpc_requests`
-
-Garage implements rate limiting for RPC requests: no more than
-`max_concurrent_rpc_requests` concurrent outbound RPC requests will be made
-by a Garage node (additionnal requests will be put in a waiting queue).
-
 #### `sled_cache_capacity`
 
 This parameter can be used to tune the capacity of the cache used by
@@ -143,21 +190,6 @@ of a power outage (though this should not matter much as data is replicated on o
 nodes). The default value, 2000ms, should be appropriate for most use cases.
 
 
-## The `[rpc_tls]` section
-
-This section should be used to configure the TLS certificates used to encrypt
-intra-cluster traffic (RPC traffic). The following parameters should be set:
-
-- `ca_cert`: the certificate of the CA that is allowed to sign individual node certificates
-- `node_cert`: the node certificate for the current node
-- `node_key`: the key associated with the node certificate
-
-Note tha several nodes may use the same node certificate, as long as it is signed
-by the CA.
-
-If this section is absent, TLS is not used to encrypt intra-cluster traffic.
-
-
 ## The `[s3_api]` section
 
 #### `api_bind_addr`
@@ -171,6 +203,15 @@ Garage will accept S3 API calls that are targetted to the S3 region defined here
 API calls targetted to other regions will fail with a AuthorizationHeaderMalformed error
 message that redirects the client to the correct region.
 
+#### `root_domain`
+
+The optionnal suffix to access bucket using vhost-style in addition to path-style request.
+Note path-style requests are always enabled, whether or not vhost-style is configured.
+Configuring vhost-style S3 required a wildcard DNS entry, and possibly a wildcard TLS certificate,
+but might be required by softwares not supporting path-style requests.
+
+If `root_domain` is `s3.garage.eu`, a bucket called `my-bucket` can be interacted with
+using the hostname `my-bucket.s3.garage.eu`.
 
 ## The `[s3_web]` section
 

doc/book/src/reference_manual/layout.md

@@ -0,0 +1,74 @@
+# Creating and updating a cluster layout
+
+The cluster layout in Garage is a table that assigns to each node a role in
+the cluster. The role of a node in Garage can either be a storage node with
+a certain capacity, or a gateway node that does not store data and is only
+used as an API entry point for faster cluster access.
+An introduction to building cluster layouts can be found in the [production deployment](/cookbook/real_world.md) page.
+
+## How cluster layouts work in Garage
+
+In Garage, a cluster layout is composed of the following components:
+
+- a table of roles assigned to nodes
+- a version number
+
+Garage nodes will always use the cluster layout with the highest version number.
+
+Garage nodes also maintain and synchronize between them a set of proposed role
+changes that haven't yet been applied. These changes will be applied (or
+canceled) in the next version of the layout
+
+The following commands insert modifications to the set of proposed role changes
+for the next layout version (but they do not create the new layout immediately):
+
+```bash
+garage layout assign [...]
+garage layout remove [...]
+```
+
+The following command can be used to inspect the layout that is currently set in the cluster
+and the changes proposed for the next layout version, if any:
+
+```bash
+garage layout show
+```
+
+The following commands create a new layout with the specified version number,
+that either takes into account the proposed changes or cancels them:
+
+```bash
+garage layout apply --version <new_version_number>
+garage layout revert --version <new_version_number>
+```
+
+The version number of the new layout to create must be 1 + the version number
+of the previous layout that existed in the cluster.  The `apply` and `revert`
+commands will fail otherwise.
+
+## Warnings about Garage cluster layout management
+
+**Warning: never make several calls to `garage layout apply` or `garage layout
+revert` with the same value of the `--version` flag. Doing so can lead to the
+creation of several different layouts with the same version number, in which
+case your Garage cluster will become inconsistent until fixed.** If a call to
+`garage layout apply` or `garage layout revert` has failed and `garage layout
+show` indicates that a new layout with the given version number has not been
+set in the cluster, then it is fine to call the command again with the same
+version number.
+
+If you are using the `garage` CLI by typing individual commands in your
+shell, you shouldn't have much issues as long as you run commands one after
+the other and take care of checking the output of `garage layout show`
+before applying any changes.
+
+If you are using the `garage` CLI to script layout changes, follow the following recommendations:
+
+- Make all of your `garage` CLI calls to the same RPC host. Do not use the
+  `garage` CLI to connect to individual nodes to send them each a piece of the
+  layout changes you are making, as the changes propagate asynchronously
+  between nodes and might not all be taken into account at the time when the
+  new layout is applied.
+
+- **Only call `garage layout apply` once**, and call it **strictly after** all
+  of the `layout assign` and `layout remove` commands have returned.

doc/book/src/reference_manual/s3_compatibility.md

@@ -5,94 +5,60 @@
 Implemented:
 
 - path-style URLs (`garage.tld/bucket/key`)
+- vhost-style URLs (`bucket.garage.tld/key`)
 - putting and getting objects in buckets
 - multipart uploads
 - listing objects
-- access control on a per-key-per-bucket basis
+- access control on a per-access-key-per-bucket basis
+- CORS headers on web endpoint
 
 Not implemented:
 
-- vhost-style URLs (`bucket.garage.tld/key`)
 - object-level ACL
-- object versioning
+- [object versioning](https://git.deuxfleurs.fr/Deuxfleurs/garage/issues/166)
 - encryption
 - most `x-amz-` headers
 
 
 ## Endpoint implementation
 
-All APIs that are not mentionned are not implemented and will return a 400 bad request.
+All APIs that are not mentionned are not implemented and will return a 501 Not Implemented.
 
-#### AbortMultipartUpload
+| Endpoint                     | Status                           |
+|------------------------------|----------------------------------|
+| AbortMultipartUpload         | Implemented                      |
+| CompleteMultipartUpload      | Implemented                      |
+| CopyObject                   | Implemented                      |
+| CreateBucket                 | Implemented                      |
+| CreateMultipartUpload        | Implemented                      |
+| DeleteBucket                 | Implemented                      |
+| DeleteBucketCors             | Implemented                      |
+| DeleteBucketWebsite          | Implemented                      |
+| DeleteObject                 | Implemented                      |
+| DeleteObjects                | Implemented                      |
+| GetBucketCors                | Implemented                      |
+| GetBucketLocation            | Implemented                      |
+| GetBucketVersioning          | Stub (see below)                 |
+| GetBucketWebsite             | Implemented                      |
+| GetObject                    | Implemented                      |
+| HeadBucket                   | Implemented                      |
+| HeadObject                   | Implemented                      |
+| ListBuckets                  | Implemented                      |
+| ListObjects                  | Implemented, bugs? (see below)   |
+| ListObjectsV2                | Implemented                      |
+| ListMultipartUpload          | Implemented                      |
+| ListParts                    | Missing                          |
+| PutObject                    | Implemented                      |
+| PutBucketCors                | Implemented                      |
+| PutBucketWebsite             | Partially implemented (see below)|
+| UploadPart                   | Implemented                      |
+| UploadPartCopy               | Implemented                      |
 
-Implemented.
 
-#### CompleteMultipartUpload
-
-Implemented badly. Garage will not check that all the parts stored correspond to the list given by the client in the request body. This means that the multipart upload might be completed with an invalid size. This is a bug and will be fixed.
-
-#### CopyObject
-
-Implemented.
-
-#### CreateBucket
-
-Garage does not accept creating buckets or giving access using API calls, it has to be done using the CLI tools. CreateBucket will return a 200 if the bucket exists and user has write access, and a 403 Forbidden in all other cases.
-
-#### CreateMultipartUpload
-
-Implemented.
-
-#### DeleteBucket
-
-Garage does not accept deleting buckets using API calls, it has to be done using the CLI tools. This request will return a 403 Forbidden.
-
-#### DeleteObject
-
-Implemented.
-
-#### DeleteObjects
-
-Implemented.
-
-#### GetBucketLocation
-
-Implemented.
-
-#### GetBucketVersioning
-
-Stub implementation (Garage does not yet support versionning so this always returns
+- **GetBucketVersioning:** Stub implementation (Garage does not yet support versionning so this always returns
 "versionning not enabled").
 
-#### GetObject
+- **ListObjects:** Implemented, but there isn't a very good specification of what `encoding-type=url` covers so there might be some encoding bugs. In our implementation the url-encoded fields are in the same in ListObjects as they are in ListObjectsV2.
 
-Implemented.
-
-#### HeadBucket
-
-Implemented.
-
-#### HeadObject
-
-Implemented.
-
-#### ListBuckets
-
-Implemented.
-
-#### ListObjects
-
-Implemented, but there isn't a very good specification of what `encoding-type=url` covers so there might be some encoding bugs. In our implementation the url-encoded fields are in the same in ListObjects as they are in ListObjectsV2.
-
-#### ListObjectsV2
-
-Implemented.
-
-#### PutObject
-
-Implemented.
-
-#### UploadPart
-
-Implemented.
+- **PutBucketWebsite:** Implemented, but only stores the index document suffix and the error document path. Redirects are not supported.
 

doc/book/src/working_documents/compatibility_target.md

@@ -0,0 +1,106 @@
+# S3 compatibility target
+
+If there is a specific S3 functionnality you have a need for, feel free to open
+a PR to put the corresponding endpoints higher in the list.  Please explain
+your motivations for doing so in the PR message.
+
+| Priority                   | Endpoints  |
+| -------------------------- | --------- |
+| **S-tier** (high priority) | |
+| 							 | HeadBucket |
+| 							 | GetBucketLocation |
+| 							 | CreateBucket |
+| 							 | DeleteBucket |
+| 							 | ListBuckets |
+| 							 | ListObjects |
+| 							 | ListObjectsV2 |
+| 							 | HeadObject |
+| 							 | GetObject |
+| 							 | PutObject |
+| 							 | CopyObject |
+| 							 | DeleteObject |
+| 							 | DeleteObjects |
+| 							 | CreateMultipartUpload |
+| 							 | CompleteMultipartUpload |
+| 							 | AbortMultipartUpload |
+| 							 | UploadPart |
+| 							 | ListMultipartUploads |
+| 							 | ListParts |
+| **A-tier**                 | |
+| 							 | GetBucketCors |
+| 							 | PutBucketCors |
+| 							 | DeleteBucketCors |
+| 							 | UploadPartCopy |
+| 							 | GetBucketWebsite |
+| 							 | PutBucketWebsite |
+| 							 | DeleteBucketWebsite |
+| 							 | [PostObject](https://docs.aws.amazon.com/AmazonS3/latest/API/RESTObjectPOST.html) |
+| ~~~~~~~~~~~~~~~~~~~~~~~~~~ | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
+| **B-tier** | |
+| 							 | GetBucketAcl |
+| 							 | PutBucketAcl |
+| 							 | GetObjectLockConfiguration |
+| 							 | PutObjectLockConfiguration |
+| 							 | GetObjectRetention |
+| 							 | PutObjectRetention |
+| 							 | GetObjectLegalHold |
+| 							 | PutObjectLegalHold |
+| **C-tier** | |
+| 							 | GetBucketVersioning |
+| 							 | PutBucketVersioning |
+| 							 | ListObjectVersions |
+| 							 | GetObjectAcl |
+| 							 | PutObjectAcl |
+| 							 | GetBucketLifecycleConfiguration |
+| 							 | PutBucketLifecycleConfiguration |
+| 							 | DeleteBucketLifecycle |
+| **garbage-tier**   | |
+| 							 | DeleteBucketEncryption |
+| 							 | DeleteBucketAnalyticsConfiguration |
+| 							 | DeleteBucketIntelligentTieringConfiguration |
+| 							 | DeleteBucketInventoryConfiguration |
+| 							 | DeleteBucketMetricsConfiguration |
+| 							 | DeleteBucketOwnershipControls |
+| 							 | DeleteBucketPolicy |
+| 							 | DeleteBucketReplication |
+| 							 | DeleteBucketTagging |
+| 							 | DeleteObjectTagging |
+| 							 | DeletePublicAccessBlock |
+| 							 | GetBucketAccelerateConfiguration |
+| 							 | GetBucketAnalyticsConfiguration |
+| 							 | GetBucketEncryption |
+| 							 | GetBucketIntelligentTieringConfiguration |
+| 							 | GetBucketInventoryConfiguration |
+| 							 | GetBucketLogging |
+| 							 | GetBucketMetricsConfiguration |
+| 							 | GetBucketNotificationConfiguration |
+| 							 | GetBucketOwnershipControls |
+| 							 | GetBucketPolicy |
+| 							 | GetBucketPolicyStatus |
+| 							 | GetBucketReplication |
+| 							 | GetBucketRequestPayment |
+| 							 | GetBucketTagging |
+| 							 | GetObjectTagging |
+| 							 | GetObjectTorrent |
+| 							 | GetPublicAccessBlock |
+| 							 | ListBucketAnalyticsConfigurations |
+| 							 | ListBucketIntelligentTieringConfigurations |
+| 							 | ListBucketInventoryConfigurations |
+| 							 | ListBucketMetricsConfigurations |
+| 							 | PutBucketAccelerateConfiguration |
+| 							 | PutBucketAnalyticsConfiguration |
+| 							 | PutBucketEncryption |
+| 							 | PutBucketIntelligentTieringConfiguration |
+| 							 | PutBucketInventoryConfiguration |
+| 							 | PutBucketLogging |
+| 							 | PutBucketMetricsConfiguration |
+| 							 | PutBucketNotificationConfiguration |
+| 							 | PutBucketOwnershipControls |
+| 							 | PutBucketPolicy |
+| 							 | PutBucketReplication |
+| 							 | PutBucketRequestPayment |
+| 							 | PutBucketTagging |
+| 							 | PutObjectTagging |
+| 							 | PutPublicAccessBlock |
+| 							 | RestoreObject |
+| 							 | SelectObjectContent |

doc/book/src/working_documents/design_draft.md

@@ -0,0 +1,162 @@
+# Design draft
+
+**WARNING: this documentation is a design draft which was written before Garage's actual implementation.
+The general principle are similar, but details have not been updated.**
+
+
+#### Modules
+
+- `membership/`: configuration, membership management (gossip of node's presence and status), ring generation --> what about Serf (used by Consul/Nomad) : https://www.serf.io/? Seems a huge library with many features so maybe overkill/hard to integrate
+- `metadata/`: metadata management
+- `blocks/`: block management, writing, GC and rebalancing
+- `internal/`: server to server communication (HTTP server and client that reuses connections, TLS if we want, etc)
+- `api/`: S3 API
+- `web/`: web management interface
+
+#### Metadata tables
+
+**Objects:**
+
+- *Hash key:* Bucket name (string)
+- *Sort key:* Object key (string)
+- *Sort key:* Version timestamp (int)
+- *Sort key:* Version UUID (string)
+- Complete: bool
+- Inline: bool, true for objects < threshold (say 1024)
+- Object size (int)
+- Mime type (string)
+- Data for inlined objects (blob)
+- Hash of first block otherwise (string)
+
+*Having only a hash key on the bucket name will lead to storing all file entries of this table for a specific bucket on a single node. At the same time, it is the only way I see to rapidly being able to list all bucket entries...*
+
+**Blocks:**
+
+- *Hash key:* Version UUID (string)
+- *Sort key:* Offset of block in total file (int)
+- Hash of data block (string)
+
+A version is defined by the existence of at least one entry in the blocks table for a certain version UUID.
+We must keep the following invariant: if a version exists in the blocks table, it has to be referenced in the objects table.
+We explicitly manage concurrent versions of an object: the version timestamp and version UUID columns are index columns, thus we may have several concurrent versions of an object.
+Important: before deleting an older version from the objects table, we must make sure that we did a successfull delete of the blocks of that version from the blocks table.
+
+Thus, the workflow for reading an object is as follows:
+
+1. Check permissions (LDAP)
+2. Read entry in object table. If data is inline, we have its data, stop here.
+   -> if several versions, take newest one and launch deletion of old ones in background
+3. Read first block from cluster. If size <= 1 block, stop here.
+4. Simultaneously with previous step, if size > 1 block: query the Blocks table for the IDs of the next blocks
+5. Read subsequent blocks from cluster
+
+Workflow for PUT:
+
+1. Check write permission (LDAP)
+2. Select a new version UUID
+3. Write a preliminary entry for the new version in the objects table with complete = false
+4. Send blocks to cluster and write entries in the blocks table
+5. Update the version with complete = true and all of the accurate information (size, etc)
+6. Return success to the user
+7. Launch a background job to check and delete older versions
+
+Workflow for DELETE:
+
+1. Check write permission (LDAP)
+2. Get current version (or versions) in object table
+3. Do the deletion of those versions NOT IN A BACKGROUND JOB THIS TIME
+4. Return succes to the user if we were able to delete blocks from the blocks table and entries from the object table
+
+To delete a version:
+
+1. List the blocks from Cassandra
+2. For each block, delete it from cluster. Don't care if some deletions fail, we can do GC.
+3. Delete all of the blocks from the blocks table
+4. Finally, delete the version from the objects table
+
+Known issue: if someone is reading from a version that we want to delete and the object is big, the read might be interrupted. I think it is ok to leave it like this, we just cut the connection if data disappears during a read.
+
+("Soit P un problème, on s'en fout est une solution à ce problème")
+
+#### Block storage on disk
+
+**Blocks themselves:**
+
+- file path = /blobs/(first 3 hex digits of hash)/(rest of hash)
+
+**Reverse index for GC & other block-level metadata:**
+
+- file path = /meta/(first 3 hex digits of hash)/(rest of hash)
+- map block hash -> set of version UUIDs where it is referenced
+
+Usefull metadata:
+
+- list of versions that reference this block in the Casandra table, so that we can do GC by checking in Cassandra that the lines still exist
+- list of other nodes that we know have acknowledged a write of this block, usefull in the rebalancing algorithm
+
+Write strategy: have a single thread that does all write IO so that it is serialized (or have several threads that manage independent parts of the hash space). When writing a blob, write it to a temporary file, close, then rename so that a concurrent read gets a consistent result (either not found or found with whole content).
+
+Read strategy: the only read operation is get(hash) that returns either the data or not found (can do a corruption check as well and return corrupted state if it is the case). Can be done concurrently with writes.
+
+**Internal API:**
+
+- get(block hash) -> ok+data/not found/corrupted
+- put(block hash & data, version uuid + offset) -> ok/error
+- put with no data(block hash, version uuid + offset) -> ok/not found plz send data/error
+- delete(block hash, version uuid + offset) -> ok/error
+
+GC: when last ref is deleted, delete block.
+Long GC procedure: check in Cassandra that version UUIDs still exist and references this block.
+
+Rebalancing: takes as argument the list of newly added nodes.
+
+- List all blocks that we have. For each block:
+- If it hits a newly introduced node, send it to them.
+  Use put with no data first to check if it has to be sent to them already or not.
+  Use a random listing order to avoid race conditions (they do no harm but we might have two nodes sending the same thing at the same time thus wasting time).
+- If it doesn't hit us anymore, delete it and its reference list.
+
+Only one balancing can be running at a same time. It can be restarted at the beginning with new parameters.
+
+#### Membership management
+
+Two sets of nodes:
+
+- set of nodes from which a ping was recently received, with status: number of stored blocks, request counters, error counters, GC%, rebalancing%
+  (eviction from this set after say 30 seconds without ping)
+- set of nodes that are part of the system, explicitly modified by the operator using the web UI (persisted to disk),
+  is a CRDT using a version number for the value of the whole set
+
+Thus, three states for nodes:
+
+- healthy: in both sets
+- missing: not pingable but part of desired cluster
+- unused/draining: currently present but not part of the desired cluster, empty = if contains nothing, draining = if still contains some blocks
+
+Membership messages between nodes:
+
+- ping with current state + hash of current membership info -> reply with same info
+- send&get back membership info (the ids of nodes that are in the two sets): used when no local membership change in a long time and membership info hash discrepancy detected with first message (passive membership fixing with full CRDT gossip)
+- inform of newly pingable node(s) -> no result, when receive new info repeat to all (reliable broadcast)
+- inform of operator membership change -> no result, when receive new info repeat to all (reliable broadcast)
+
+Ring: generated from the desired set of nodes, however when doing read/writes on the ring, skip nodes that are known to be not pingable.
+The tokens are generated in a deterministic fashion from node IDs (hash of node id + token number from 1 to K).
+Number K of tokens per node: decided by the operator & stored in the operator's list of nodes CRDT. Default value proposal: with node status information also broadcast disk total size and free space, and propose a default number of tokens equal to 80%Free space / 10Gb. (this is all user interface)
+
+
+#### Constants
+
+- Block size: around 1MB ? --> Exoscale use 16MB chunks
+- Number of tokens in the hash ring: one every 10Gb of allocated storage
+- Threshold for storing data directly in Cassandra objects table: 1kb bytes (maybe up to 4kb?)
+- Ping timeout (time after which a node is registered as unresponsive/missing): 30 seconds
+- Ping interval: 10 seconds
+- ??
+
+#### Links
+
+- CDC: <https://www.usenix.org/system/files/conference/atc16/atc16-paper-xia.pdf>
+- Erasure coding: <http://web.eecs.utk.edu/~jplank/plank/papers/CS-08-627.html>
+- [Openstack Storage Concepts](https://docs.openstack.org/arch-design/design-storage/design-storage-concepts.html)
+- [RADOS](https://ceph.com/wp-content/uploads/2016/08/weil-rados-pdsw07.pdf)

doc/book/src/working_documents/load_balancing.md

@@ -1,5 +1,7 @@
 # Load Balancing Data (planned for version 0.2)
 
+**This is being yet improved in release 0.5. The working document has not been updated yet, it still only applies to Garage 0.2 through 0.4.**
+
 I have conducted a quick study of different methods to load-balance data over different Garage nodes using consistent hashing.
 
 ## Requirements

doc/book/src/working_documents/migration_04.md

@@ -0,0 +1,105 @@
+# Migrating from 0.3 to 0.4
+
+**Migrating from 0.3 to 0.4 is unsupported. This document is only intended to
+document the process internally for the Deuxfleurs cluster where we have to do
+it. Do not try it yourself, you will lose your data and we will not help you.**
+
+**Migrating from 0.2 to 0.4 will break everything for sure. Never try it.**
+
+The internal data format of Garage hasn't changed much between 0.3 and 0.4.
+The Sled database is still the same, and the data directory as well.
+
+The following has changed, all in the meta directory:
+
+- `node_id` in 0.3 contains the identifier of the current node. In 0.4, this
+  file does nothing and should be deleted. It is replaced by `node_key` (the
+  secret key) and `node_key.pub` (the associated public key). A node's
+  identifier on the ring is its public key.
+
+- `peer_info` in 0.3 contains the list of peers saved automatically by Garage.
+  The format has changed and it is now stored in `peer_list` (`peer_info`
+  should be deleted).
+
+When migrating, all node identifiers will change. This also means that the
+affectation of data partitions on the ring will change, and lots of data will
+have to be rebalanced.
+
+- If your cluster has only 3 nodes, all nodes store everything, therefore nothing has to be rebalanced.
+
+- If your cluster has only 4 nodes, for any partition there will always be at
+  least 2 nodes that stored data before that still store it after. Therefore
+  the migration should in theory be transparent and Garage should continue to
+  work during the rebalance.
+
+- If your cluster has 5 or more nodes, data will disappear during the
+  migration. Do not migrate (fortunately we don't have this scenario at
+  Deuxfleurs), or if you do, make Garage unavailable until things stabilize
+  (disable web and api access).
+
+
+The migration steps are as follows:
+
+1. Prepare a new configuration file for 0.4. For each node, point to the same
+   meta and data directories as Garage 0.3. Basically, the things that change
+   are the following:
+
+  - No more `rpc_tls` section
+  - You have to generate a shared `rpc_secret` and put it in all config files
+  - `bootstrap_peers` has a different syntax as it has to contain node keys.
+    Leave it empty and use `garage node-id` and `garage node connect` instead (new features of 0.4)
+  - put the publicly accessible RPC address of your node in `rpc_public_addr` if possible (its optional but recommended)
+  - If you are using Consul, change the `consul_service_name` to NOT be the name advertised by Nomad.
+    Now Garage is responsible for advertising its own service itself.
+
+2. Disable api and web access for some time (Garage does not support disabling
+   these endpoints but you can change the port number or stop your reverse
+   proxy for instance).
+
+3. Do `garage repair -a --yes tables` and `garage repair -a --yes blocks`,
+   check the logs and check that all data seems to be synced correctly between
+   nodes.
+
+4. Save somewhere the output of `garage status`. We will need this to remember
+   how to reconfigure nodes in 0.4.
+
+5. Turn off Garage 0.3
+
+6. Backup metadata folders if you can (i.e. if you have space to do it
+   somewhere). Backuping data folders could also be usefull but that's much
+   harder to do. If your filesystem supports snapshots, this could be a good
+   time to use them.
+
+7. Turn on Garage 0.4
+
+8. At this point, running `garage status` should indicate that all nodes of the
+   previous cluster are "unavailable". The nodes have new identifiers that
+   should appear in healthy nodes once they can talk to one another (use
+   `garage node connect` if necessary`). They should have NO ROLE ASSIGNED at
+   the moment.
+
+9. Prepare a script with several `garage node configure` commands that replace
+   each of the v0.3 node ID with the corresponding v0.4 node ID, with the same
+   zone/tag/capacity. For example if your node `drosera` had identifier `c24e`
+   before and now has identifier `789a`, and it was configured with capacity
+   `2` in zone `dc1`, put the following command in your script:
+
+```bash
+garage node configure 789a -z dc1 -c 2 -t drosera --replace c24e
+```
+
+10. Run your reconfiguration script. Check that the new output of `garage
+	status` contains the correct node IDs with the correct values for capacity
+	and zone. Old nodes should no longer be mentioned.
+
+11. If your cluster has 4 nodes or less, and you are feeling adventurous, you
+	can reenable Web and API access now. Things will probably work.
+
+12. Garage might already be resyncing stuff. Issue a `garage repair -a --yes
+	tables` and `garage repair -a --yes blocks` to force it to do so.
+
+13. Wait for resyncing activity to stop in the logs. Do steps 12 and 13 two or
+	three times, until you see that when you issue the repair commands, nothing
+	gets resynced any longer.
+
+14. Your upgraded cluster should be in a working state. Re-enable API and Web
+	access and check that everything went well.

doc/book/src/working_documents/migration_06.md

@@ -0,0 +1,46 @@
+# Migrating from 0.5 to 0.6
+
+**This guide explains how to migrate to 0.6 if you have an existing 0.5 cluster.
+We don't recommend trying to migrate directly from 0.4 or older to 0.6.**
+
+**We make no guarantee that this migration will work perfectly:
+back up all your data before attempting it!**
+
+Garage v0.6 (not yet released) introduces a new data model for buckets,
+that allows buckets to have many names (aliases).
+Buckets can also have "private" aliases (called local aliases),
+which are only visible when using a certain access key.
+
+This new data model means that the metadata tables have changed quite a bit in structure,
+and a manual migration step is required.
+
+The migration steps are as follows:
+
+1. Disable api and web access for some time (Garage does not support disabling
+   these endpoints but you can change the port number or stop your reverse
+   proxy for instance).
+
+2. Do `garage repair -a --yes tables` and `garage repair -a --yes blocks`,
+   check the logs and check that all data seems to be synced correctly between
+   nodes.
+
+4. Turn off Garage 0.5
+
+5. **Backup your metadata folders!!**
+
+6. Turn on Garage 0.6
+
+7. At this point, `garage bucket list` should indicate that no buckets are present
+   in the cluster. `garage key list` should show all of the previously existing
+   access key, however these keys should not have any permissions to access buckets.
+
+8. Run `garage migrate buckets050`: this will populate the new bucket table with
+   the buckets that existed previously. This will also give access to API keys
+   as it was before.
+
+9. Check that all your buckets indeed appear in `garage bucket list`, and that
+   keys have the proper access flags set. If that is not the case, revert
+   everything and file a bug!
+
+10. Your upgraded cluster should be in a working state. Re-enable API and Web
+    access and check that everything went well.

doc/talks/2021-09-13-ngi-kickoff/.gitignore

@@ -0,0 +1,12 @@
+*
+
+!img
+
+!.gitignore
+!*.svg
+!*.png
+!*.jpg
+!*.tex
+!Makefile
+!.gitignore
+!talk.pdf

doc/talks/2021-09-13-ngi-kickoff/Makefile

@@ -0,0 +1,3 @@
+talk.pdf: talk.tex
+	pdflatex talk.tex
+

doc/talks/2021-09-13-ngi-kickoff/talk.tex

@@ -0,0 +1,147 @@
+%\nonstopmode
+\documentclass[aspectratio=169]{beamer}
+\usepackage[utf8]{inputenc}
+% \usepackage[frenchb]{babel}
+\usepackage{amsmath}
+\usepackage{mathtools}
+\usepackage{breqn}
+\usepackage{multirow}
+\usetheme{boxes}
+\usepackage{graphicx}
+%\useoutertheme[footline=authortitle,subsection=false]{miniframes}
+
+\beamertemplatenavigationsymbolsempty
+
+\usepackage{tabu}
+\usepackage{multicol}
+\usepackage{vwcol}
+\usepackage{stmaryrd}
+\usepackage{graphicx}
+
+\usepackage[normalem]{ulem}
+
+\title{Presentation of the Garage project}
+\subtitle{NGI pointer kickoff meeting}
+\author{Deuxfleurs Association}
+\date{2021-09-13}
+
+\begin{document}
+
+\begin{frame}
+	\centering
+	\includegraphics[width=.3\linewidth]{../../sticker/Garage.pdf}
+	\vspace{1em}
+
+	{\large\bf Deuxfleurs Association}
+	\vspace{1em}
+
+	\url{https://deuxfleurs.fr/}
+
+	\url{https://garagehq.deuxfleurs.fr/}
+
+	Matrix channel: \texttt{\#garage:deuxfleurs.fr}
+\end{frame}
+
+\begin{frame}
+	\frametitle{Our objective at Deuxfleurs}
+	
+	\begin{center}
+		\textbf{Promote self-hosting and small-scale hosting\\
+			as an alternative to large cloud providers}
+	\end{center}
+	\vspace{2em}
+	\visible<2->{
+		Why is it hard?
+	}
+	\visible<3->{
+		\vspace{2em}
+		\begin{center}
+			\textbf{\underline{Resilience}}\\
+			{\footnotesize (we want good uptime/availability with low supervision)}
+		\end{center}
+	}
+\end{frame}
+
+\begin{frame}
+	\frametitle{How to be resilient (the hard way)}
+
+	Entreprise-grade systems typically employ:
+	\vspace{1em}
+	\begin{itemize}
+		\item Redundant Internet connections
+		\item Redundant electricity
+		\item UPSes
+		\item RAID
+		\item ... 
+	\end{itemize}
+	\vspace{1em}
+	$\to$ it's costly and only worth it at DC scale
+\end{frame}
+
+\begin{frame}
+	\frametitle{How to be resilient (the \underline{\textbf{cheap}} way)}
+
+	Instead, we use:
+	\vspace{1em}
+	\begin{itemize}
+		\item Commodity hardware (e.g. old desktop PCs)
+			\vspace{.5em}
+		\item<2-> Commodity Internet (e.g. FTTB, FTTH) and electricity
+			\vspace{.5em}
+		\item<3-> \textbf{Geographical redundancy} (multi-site replication)
+			\vspace{.5em}
+		\item<4-> \textbf{Fault-tolerant distributed algorithms}
+	\end{itemize}
+	\vspace{1em}
+	\visible<5->{
+		\centering
+		\underline{\textbf{This is how we build Garage.}}
+	}
+\end{frame}
+
+\begin{frame}
+	\frametitle{But what is Garage, exactly?}
+	\textbf{Garage is a self-hosted drop-in replacement for the Amazon S3 object store}\\
+	\vspace{.5em}
+	that implements resilience through geographical redundancy on commodity hardware
+
+	\vspace{1em}
+	\visible<2->{
+		\begin{center}
+				Current status: technical preview\\
+				Our goal: release a stable v1.0
+		\end{center}
+	}
+
+	\vspace{1em}
+	\visible<3->{
+		\textbf{Comming up next: an e-mail IMAP inbox based on the same principles}
+	}
+
+	\vspace{1em}
+	\visible<4->{
+		\begin{center}
+				Current status: just an idea\\
+				Our goal: at least a PoC, maybe more
+		\end{center}
+	}
+\end{frame}
+
+\begin{frame}
+	\centering
+	\includegraphics[width=.3\linewidth]{../../sticker/Garage.pdf}
+	\vspace{1em}
+
+	{\large\bf Deuxfleurs Association}
+	\vspace{1em}
+
+	\url{https://deuxfleurs.fr/}
+
+	\url{https://garagehq.deuxfleurs.fr/}
+
+	Matrix channel: \texttt{\#garage:deuxfleurs.fr}
+\end{frame}
+
+\end{document}
+
+%% vim: set ts=4 sw=4 tw=0 noet spelllang=fr :

genkeys.sh

@@ -1,83 +0,0 @@
-#!/bin/bash
-
-set -xe
-
-cd $(dirname $0)
-
-mkdir -p pki
-cd pki
-
-# Create a certificate authority that both the client side and the server side of
-# the RPC protocol will use to authenticate the other side.
-if [ ! -f garage-ca.key ]; then
-    echo "Generating Garage CA keys..."
-	openssl genpkey -algorithm ED25519 -out garage-ca.key
-	openssl req -x509 -new -nodes -key garage-ca.key -sha256 -days 3650 -out garage-ca.crt -subj "/C=FR/O=Garage"
-fi
-
-
-# Generate a certificate that can be used either as a server certificate
-# or a client certificate. This is what the RPC client and server will use
-# to prove that they are authenticated by the CA.
-if [ ! -f garage.crt ]; then
-	echo "Generating Garage agent keys..."
-	if [ ! -f garage.key ]; then
-		openssl genpkey -algorithm ED25519 -out garage.key
-	fi
-	openssl req -new -sha256 -key garage.key -subj "/C=FR/O=Garage/CN=garage" \
-		-out garage.csr
-	openssl req -in garage.csr -noout -text
-	openssl x509 -req -in garage.csr \
-		-extensions v3_req \
-		-extfile <(cat <<EOF
-[req]
-distinguished_name = req_distinguished_name
-req_extensions = v3_req
-prompt = no
-
-[req_distinguished_name]
-C = FR
-O = Garage
-CN = garage
-
-[v3_req]
-keyUsage = keyEncipherment, dataEncipherment
-extendedKeyUsage = serverAuth, clientAuth
-subjectAltName = @alt_names
-[alt_names]
-DNS.1 = garage
-EOF
-) \
-		-CA garage-ca.crt -CAkey garage-ca.key -CAcreateserial \
-		-out garage.crt -days 365
-fi
-
-# Client-only certificate used for the CLI
-if [ ! -f garage-client.crt ]; then
-	echo "Generating Garage client keys..."
-	if [ ! -f garage-client.key ]; then
-		openssl genpkey -algorithm ED25519 -out garage-client.key
-	fi
-	openssl req -new -sha256 -key garage-client.key -subj "/C=FR/O=Garage" \
-		-out garage-client.csr
-	openssl req -in garage-client.csr -noout -text
-	openssl x509 -req -in garage-client.csr \
-		-extensions v3_req \
-		-extfile <(cat <<EOF
-[req]
-distinguished_name = req_distinguished_name
-req_extensions = v3_req
-prompt = no
-
-[req_distinguished_name]
-C = FR
-O = Garage
-
-[v3_req]
-keyUsage = keyEncipherment, dataEncipherment
-extendedKeyUsage = clientAuth
-EOF
-) \
-		-CA garage-ca.crt -CAkey garage-ca.key -CAcreateserial \
-		-out garage-client.crt -days 365
-fi

make_docker.sh

@@ -1,20 +0,0 @@
-#!/bin/sh
-
-BIN=target/release/garage
-DOCKER=lxpz/garage_amd64
-
-TAG=$1
-if [ -z "$1" ]; then
-	echo "Usage: $0 <tag>"
-	exit 1
-fi
-
-RUSTFLAGS="-C link-arg=-fuse-ld=lld -C target-cpu=x86-64 -C target-feature=+sse2" cargo build --release --no-default-features
-cp $BIN $BIN.stripped
-strip $BIN.stripped
-
-docker pull archlinux:latest
-docker build -t $DOCKER:$TAG .
-docker push $DOCKER:$TAG
-docker tag $DOCKER:$TAG $DOCKER:latest
-docker push $DOCKER:latest

nix/build_index.nix

@@ -0,0 +1,146 @@
+{
+  path ? "/../aws-list.txt",
+}:
+
+with import ./common.nix;
+let 
+  pkgs = import pkgsSrc {};
+  lib = pkgs.lib;
+
+  /* Converts a key list and a value list to a set
+
+     Example:
+       listToSet [ "name" "version" ] [ "latex" "3.14" ]
+       => { name = "latex"; version = "3.14"; }
+  */
+  listToSet = keys: values: 
+    builtins.listToAttrs 
+      (lib.zipListsWith 
+        (a: b: { name = a; value = b; }) 
+        keys
+        values);
+
+  /* Says if datetime a is more recent than datetime b
+
+    Example:
+      cmpDate { date = "2021-09-10"; time = "22:12:15"; } { date = "2021-02-03"; time = "23:54:12"; }
+      => true
+  */
+  cmpDate = a: b: 
+    let da = (builtins.head a.builds).date;
+        db = (builtins.head b.builds).date;
+    in 
+      if da == db then  (builtins.head a.builds).time > (builtins.head b.builds).time
+      else da > db;
+
+  /* Pretty platforms */
+  prettyPlatform = name:
+    if name == "aarch64-unknown-linux-musl" then "linux/arm64"
+    else if name == "armv6l-unknown-linux-musleabihf" then "linux/arm"
+    else if name == "x86_64-unknown-linux-musl" then "linux/amd64"
+    else if name == "i686-unknown-linux-musl" then "linux/386"
+    else name;
+
+  /* Parsing */
+  list = builtins.readFile (./. + path);
+  entries = lib.splitString "\n" list;
+
+  elems = builtins.filter 
+    (e: (builtins.length e) == 4) 
+    (map 
+      (x: builtins.filter (e: e != "") (lib.splitString " " x)) 
+      entries);
+
+  keys = ["date" "time" "size" "path"];
+  parsed = map (entry: listToSet keys entry) elems;
+
+  subkeys = ["root" "version" "platform" "binary" ];
+  builds = map (entry: entry // listToSet subkeys (lib.splitString "/" entry.path)) parsed;
+
+  /* Aggregation */
+  builds_per_version = lib.foldl (acc: v: acc // { ${v.version} = if builtins.hasAttr v.version acc then acc.${v.version} ++ [ v ] else [ v ]; }) {} builds; 
+
+  versions = builtins.attrNames builds_per_version;
+  versions_release = builtins.filter (x: builtins.match "v[0-9]+\.[0-9]+\.[0-9]+(\.[0-9]+)?" x != null) versions;
+  versions_commit = builtins.filter (x: builtins.match "[0-9a-f]{40}" x != null) versions;
+  versions_extra = lib.subtractLists (versions_release ++ versions_commit) versions;
+
+  sorted_builds = [
+    {
+      name = "Release";
+      hide = false;
+      type = "tag";
+      description = "Release builds are the official builds, they are tailored for productions and are the most tested.";
+      builds =  builtins.sort (a: b: a.version > b.version) (map (x: { version = x; builds = builtins.getAttr x builds_per_version; }) versions_release);
+    }
+    {
+      name = "Extra";
+      hide = true;
+      type = "tag";
+      description = "Extra builds are built on demand to test a specific feature or a specific need.";
+      builds = builtins.sort cmpDate (map (x: { version = x; builds = builtins.getAttr x builds_per_version; }) versions_extra);
+    }
+    {
+      name = "Development";
+      hide = true;
+      type = "commit";
+      description = "Development builds are built periodically. Use them if you want to test a specific feature that is not yet released.";
+      builds = builtins.sort cmpDate (map (x: { version = x; builds = builtins.getAttr x builds_per_version; }) versions_commit);
+    }
+  ];
+
+in
+  pkgs.writeText "index.html" ''
+<!doctype html>
+<html>
+  <head>
+    <meta charset="utf-8" />
+    <title>Garage releases</title>
+    <style>
+      html, body { margin:0; padding: 0 }
+      body { font-family: 'Helvetica', Sans; }
+      section { margin: 1rem; }
+      ul { padding:0; margin: 0.2rem }
+      li {
+        border-radius: 0.2rem;
+        display: inline; 
+        border: 2px #0b5d83 solid; 
+        padding: 0.5rem; 
+        line-height: 3rem;
+        color: #0b5d83;
+      }
+      li:hover { background-color: #0b5d83; color: #fff; }
+      li a, li a:hover { color: inherit; text-decoration: none }
+    </style>
+  </head>
+  <body>
+   ${ builtins.toString (lib.forEach sorted_builds (r: ''
+    <section>
+      <h2>${r.name} builds</h2>
+
+      <p>${r.description}</p>
+
+      ${if r.hide then "<details><summary>Show ${r.name} builds</summary>" else ""}
+      ${ builtins.toString (lib.forEach r.builds (x: ''
+        <h3> ${x.version} (${(builtins.head x.builds).date}) </h3>
+        <p>See this build on</p>
+        <p> Binaries:
+        <ul>
+        ${ builtins.toString (lib.forEach x.builds (b: ''
+          <li><a href="/${b.path}">${prettyPlatform b.platform}</a></li>
+        ''))}
+        </ul></p>
+        <p> Sources:
+        <ul>
+         <li><a href="https://git.deuxfleurs.fr/Deuxfleurs/garage/src/${r.type}/${x.version}">gitea</a></li>
+         <li><a href="https://git.deuxfleurs.fr/Deuxfleurs/garage/archive/${x.version}.zip">.zip</a></li>
+         <li><a href="https://git.deuxfleurs.fr/Deuxfleurs/garage/archive/${x.version}.tar.gz">.tar.gz</a></li>
+        </ul></p>
+      ''))  }
+      ${ if builtins.length r.builds == 0 then "<em>There is no build for this category</em>" else "" }
+      ${if r.hide then "</details>" else ""}
+    </section>
+  ''))}
+  </body>
+</html>
+''

nix/common.nix

@@ -0,0 +1,21 @@
+rec {
+  /*
+   * Fixed dependencies
+   */
+  pkgsSrc = fetchTarball {
+    # As of 2021-10-04
+    url ="https://github.com/NixOS/nixpkgs/archive/b27d18a412b071f5d7991d1648cfe78ee7afe68a.tar.gz";
+    sha256 = "1xy9zpypqfxs5gcq5dcla4bfkhxmh5nzn9dyqkr03lqycm9wg5cr";
+  };
+  cargo2nixSrc = fetchGit {
+    # As of 2021-10-06
+    url = "https://github.com/superboum/cargo2nix";
+    rev = "1364752cd784764db2ef5b1e1248727cebfae2ce";
+  };
+
+  /*
+   * Shared objects
+   */
+  cargo2nix = import cargo2nixSrc;
+  cargo2nixOverlay = import "${cargo2nixSrc}/overlay";
+}

nix/kaniko.nix

@@ -0,0 +1,23 @@
+pkgs:
+pkgs.buildGoModule rec {
+  pname = "kaniko";
+  version = "1.6.0";
+
+  src = pkgs.fetchFromGitHub {
+    owner = "GoogleContainerTools";
+    repo = "kaniko";
+    rev = "v${version}";
+    sha256 = "1fnclr556avxay6pvgw5ya3xbxfnf2gv4njq2hr4fd6fcjyslq5h";
+  };
+
+  vendorSha256 = null;
+
+  checkPhase = "true";
+
+  meta = with pkgs.lib; {
+    description = "kaniko is a tool to build container images from a Dockerfile, inside a container or Kubernetes cluster.";
+    homepage = "https://github.com/GoogleContainerTools/kaniko";
+    license = licenses.asl20;
+    platforms = platforms.linux;
+  };
+}

nix/nix.conf

@@ -0,0 +1,4 @@
+substituters = https://cache.nixos.org https://nix.web.deuxfleurs.fr
+trusted-public-keys = cache.nixos.org-1:6NCHdD59X431o0gWypbMrAURkbJ16ZPMQFGspcDShjY= nix.web.deuxfleurs.fr:eTGL6kvaQn6cDR/F9lDYUIP9nCVR/kkshYfLDJf1yKs=
+max-jobs = auto
+cores = 4

nix/toolchain.nix

@@ -0,0 +1,29 @@
+{
+  system ? builtins.currentSystem,
+}:
+
+with import ./common.nix;
+
+let
+  platforms = [
+    "x86_64-unknown-linux-musl"
+    "i686-unknown-linux-musl"
+    "aarch64-unknown-linux-musl"
+    "armv6l-unknown-linux-musleabihf"
+  ];
+  pkgsList = builtins.map (target: import pkgsSrc { 
+    inherit system; 
+    crossSystem = { config = target; };
+  }) platforms;
+  pkgsHost = import pkgsSrc {};
+  lib = pkgsHost.lib;
+  kaniko = (import ./kaniko.nix) pkgsHost;
+in 
+  lib.flatten (builtins.map (pkgs: [
+     pkgs.rustPlatform.rust.rustc
+     pkgs.rustPlatform.rust.cargo
+     pkgs.buildPackages.stdenv.cc
+  ]) pkgsList) ++ [
+    kaniko
+  ]
+

script/builder_image/Dockerfile

@@ -1,7 +0,0 @@
-FROM rust:buster
-RUN apt-get update && \
-	apt-get install --yes libsodium-dev awscli python-pip wget rclone openssl socat && \
-	    rm -rf /var/lib/apt/lists/*
-RUN wget https://dl.min.io/client/mc/release/linux-amd64/mc -O /usr/local/bin/mc && chmod +x /usr/local/bin/mc
-RUN rustup component add rustfmt clippy
-RUN pip install s3cmd

script/builder_image/Makefile

@@ -1,8 +0,0 @@
-DOCKER=lxpz/garage_builder_amd64
-
-docker:
-	docker build -t $(DOCKER):$(TAG) .
-	docker push $(DOCKER):$(TAG)
-	docker tag $(DOCKER):$(TAG) $(DOCKER):latest
-	docker push $(DOCKER):latest
-	

script/dev-bucket.sh

@@ -1,4 +1,4 @@
-#!/bin/bash
+#!/usr/bin/env bash
 
 set -ex
 
@@ -6,13 +6,14 @@ SCRIPT_FOLDER="`dirname \"$0\"`"
 REPO_FOLDER="${SCRIPT_FOLDER}/../"
 GARAGE_DEBUG="${REPO_FOLDER}/target/debug/"
 GARAGE_RELEASE="${REPO_FOLDER}/target/release/"
-PATH="${GARAGE_DEBUG}:${GARAGE_RELEASE}:$PATH"
+NIX_RELEASE="${REPO_FOLDER}/result/bin/"
+PATH="${GARAGE_DEBUG}:${GARAGE_RELEASE}:${NIX_RELEASE}:$PATH"
 
-garage bucket create eprouvette
-KEY_INFO=`garage key new --name opérateur`
+garage -c /tmp/config.1.toml bucket create eprouvette
+KEY_INFO=$(garage -c /tmp/config.1.toml key new --name opérateur)
 ACCESS_KEY=`echo $KEY_INFO|grep -Po 'GK[a-f0-9]+'`
 SECRET_KEY=`echo $KEY_INFO|grep -Po 'Secret key: [a-f0-9]+'|grep -Po '[a-f0-9]+$'`
-garage bucket allow eprouvette --read --write --key $ACCESS_KEY
+garage -c /tmp/config.1.toml bucket allow eprouvette --owner --read --write --key $ACCESS_KEY
 echo "$ACCESS_KEY $SECRET_KEY" > /tmp/garage.s3
 
 echo "Bucket s3://eprouvette created. Credentials stored in /tmp/garage.s3."

script/dev-clean.sh

@@ -1,4 +1,4 @@
-#!/bin/bash
+#!/usr/bin/env bash
 
 set -ex
 

script/dev-cluster.sh

@@ -1,4 +1,4 @@
-#!/bin/bash
+#!/usr/bin/env bash
 
 set -e
 
@@ -6,7 +6,8 @@ SCRIPT_FOLDER="`dirname \"$0\"`"
 REPO_FOLDER="${SCRIPT_FOLDER}/../"
 GARAGE_DEBUG="${REPO_FOLDER}/target/debug/"
 GARAGE_RELEASE="${REPO_FOLDER}/target/release/"
-PATH="${GARAGE_DEBUG}:${GARAGE_RELEASE}:$PATH"
+NIX_RELEASE="${REPO_FOLDER}/result/bin/"
+PATH="${GARAGE_DEBUG}:${GARAGE_RELEASE}:${NIX_RELEASE}:$PATH"
 FANCYCOLORS=("41m" "42m" "44m" "45m" "100m" "104m")
 
 export RUST_BACKTRACE=1 
@@ -16,6 +17,10 @@ MAIN_LABEL="\e[${FANCYCOLORS[0]}[main]\e[49m"
 WHICH_GARAGE=$(which garage || exit 1)
 echo -en "${MAIN_LABEL} Found garage at: ${WHICH_GARAGE}\n"
 
+NETWORK_SECRET="$(openssl rand -hex 32)"
+
+
+# <<<<<<<<< BEGIN FOR LOOP ON NODES
 for count in $(seq 1 3); do
 CONF_PATH="/tmp/config.$count.toml"
 LABEL="\e[${FANCYCOLORS[$count]}[$count]\e[49m"
@@ -25,26 +30,28 @@ block_size = 1048576			# objects are split in blocks of maximum this number of b
 metadata_dir = "/tmp/garage-meta-$count"
 data_dir = "/tmp/garage-data-$count"
 rpc_bind_addr = "0.0.0.0:$((3900+$count))"		# the port other Garage nodes will use to talk to this node
-bootstrap_peers = [
-  "127.0.0.1:3901",
-  "127.0.0.1:3902",
-  "127.0.0.1:3903"
-]
-max_concurrent_rpc_requests = 12
+rpc_public_addr = "127.0.0.1:$((3900+$count))"
+bootstrap_peers = []
 replication_mode = "3"
+rpc_secret = "$NETWORK_SECRET"
 
 [s3_api]
 api_bind_addr = "0.0.0.0:$((3910+$count))"	# the S3 API port, HTTP without TLS. Add a reverse proxy for the TLS part.
 s3_region = "garage"				# set this to anything. S3 API calls will fail if they are not made against the region set here.
+root_domain = ".s3.garage.localhost"
 
 [s3_web]
 bind_addr = "0.0.0.0:$((3920+$count))"
-root_domain = ".garage.tld"
+root_domain = ".web.garage.localhost"
 index = "index.html"
 EOF
 
 echo -en "$LABEL configuration written to $CONF_PATH\n"
 
+(garage -c /tmp/config.$count.toml server 2>&1|while read r; do echo -en "$LABEL $r\n"; done) &
+done
+# >>>>>>>>>>>>>>>> END FOR LOOP ON NODES
+
 if [ -z "$SKIP_HTTPS" ]; then
   echo -en "$LABEL Starting dummy HTTPS reverse proxy\n"
   mkdir -p /tmp/garagessl
@@ -60,13 +67,26 @@ if [ -z "$SKIP_HTTPS" ]; then
   socat openssl-listen:4443,reuseaddr,fork,cert=/tmp/garagessl/test.pem,verify=0 tcp4-connect:localhost:3911 &
 fi
 
-(garage server -c /tmp/config.$count.toml 2>&1|while read r; do echo -en "$LABEL $r\n"; done) &
+sleep 3
+# Establish connections between nodes
+for count in $(seq 1 3); do
+	NODE=$(garage -c /tmp/config.$count.toml node id -q)
+	for count2 in $(seq 1 3); do
+		garage -c /tmp/config.$count2.toml node connect $NODE
+	done
 done
 
-until garage status 2>&1|grep -q Healthy ; do 
+RETRY=120
+until garage -c /tmp/config.1.toml status 2>&1|grep -q HEALTHY ; do 
+  (( RETRY-- ))
+  if (( RETRY <= 0 )); then 
+    echo -en "${MAIN_LABEL} Garage did not start"
+    exit 1
+  fi
 	echo -en "${MAIN_LABEL} cluster starting...\n"
 	sleep 1
 done
+
 echo -en "${MAIN_LABEL} cluster started\n"
 
 wait

script/dev-configure.sh

@@ -1,4 +1,4 @@
-#!/bin/bash
+#!/usr/bin/env bash
 
 set -ex
 
@@ -6,18 +6,26 @@ SCRIPT_FOLDER="`dirname \"$0\"`"
 REPO_FOLDER="${SCRIPT_FOLDER}/../"
 GARAGE_DEBUG="${REPO_FOLDER}/target/debug/"
 GARAGE_RELEASE="${REPO_FOLDER}/target/release/"
-PATH="${GARAGE_DEBUG}:${GARAGE_RELEASE}:$PATH"
+NIX_RELEASE="${REPO_FOLDER}/result/bin/"
+PATH="${GARAGE_DEBUG}:${GARAGE_RELEASE}:${NIX_RELEASE}:$PATH"
 
 sleep 5
-until garage status 2>&1|grep -q Healthy ; do 
+RETRY=120
+until garage -c /tmp/config.1.toml status 2>&1|grep -q HEALTHY ; do 
+  (( RETRY-- ))
+  if (( RETRY <= 0 )); then
+    echo "garage did not start in time, failing."
+    exit 1
+  fi
 	echo "cluster starting..."
 	sleep 1
 done
 
-garage status \
-	| grep UNCONFIGURED \
+garage -c /tmp/config.1.toml status \
+	| grep 'NO ROLE' \
 	| grep -Po '^[0-9a-f]+' \
 	| while read id; do 
-	  garage node configure -z dc1 -c 1 $id
+	  garage -c /tmp/config.1.toml layout assign $id -z dc1 -c 1
 	done
 
+garage -c /tmp/config.1.toml layout apply --version 1

script/dev-env-aws.sh

@@ -1,5 +1,3 @@
-#!/bin/bash
-
 export AWS_ACCESS_KEY_ID=`cat /tmp/garage.s3 |cut -d' ' -f1`
 export AWS_SECRET_ACCESS_KEY=`cat /tmp/garage.s3 |cut -d' ' -f2`
 export AWS_DEFAULT_REGION='garage'

script/test-smoke.sh

@@ -1,4 +1,4 @@
-#!/bin/bash
+#!/usr/bin/env bash
 
 set -ex
 
@@ -8,27 +8,34 @@ SCRIPT_FOLDER="`dirname \"$0\"`"
 REPO_FOLDER="${SCRIPT_FOLDER}/../"
 GARAGE_DEBUG="${REPO_FOLDER}/target/debug/"
 GARAGE_RELEASE="${REPO_FOLDER}/target/release/"
-PATH="${GARAGE_DEBUG}:${GARAGE_RELEASE}:$PATH"
+NIX_RELEASE="${REPO_FOLDER}/result/bin/"
+PATH="${GARAGE_DEBUG}:${GARAGE_RELEASE}:${NIX_RELEASE}:$PATH"
+CMDOUT=/tmp/garage.cmd.tmp
 
 # @FIXME Duck is not ready for testing, we have a bug
 SKIP_DUCK=1
 
 echo "⏳ Setup"
-cargo build
 ${SCRIPT_FOLDER}/dev-clean.sh
 ${SCRIPT_FOLDER}/dev-cluster.sh > /tmp/garage.log 2>&1 &
+sleep 6
 ${SCRIPT_FOLDER}/dev-configure.sh
 ${SCRIPT_FOLDER}/dev-bucket.sh
 
 which garage
-garage status
-garage key list
-garage bucket list
+garage -c /tmp/config.1.toml status
+garage -c /tmp/config.1.toml key list
+garage -c /tmp/config.1.toml bucket list
 
 dd if=/dev/urandom of=/tmp/garage.1.rnd bs=1k count=2 # No multipart, inline storage (< INLINE_THRESHOLD = 3072 bytes)
 dd if=/dev/urandom of=/tmp/garage.2.rnd bs=1M count=5 # No multipart but file will be chunked
 dd if=/dev/urandom of=/tmp/garage.3.rnd bs=1M count=10 # by default, AWS starts using multipart at 8MB
 
+# data of lower entropy, to test compression
+dd if=/dev/urandom bs=1k count=2  | base64 -w0 > /tmp/garage.1.b64
+dd if=/dev/urandom bs=1M count=5  | base64 -w0 > /tmp/garage.2.b64
+dd if=/dev/urandom bs=1M count=10 | base64 -w0 > /tmp/garage.3.b64
+
 echo "🧪 S3 API testing..."
 
 # AWS
@@ -36,11 +43,11 @@ if [ -z "$SKIP_AWS" ]; then
   echo "🛠️ Testing with awscli"
   source ${SCRIPT_FOLDER}/dev-env-aws.sh
   aws s3 ls
-  for idx in $(seq 1 3); do
-    aws s3 cp "/tmp/garage.$idx.rnd" "s3://eprouvette/&+-é\"/garage.$idx.aws"
+  for idx in {1..3}.{rnd,b64}; do
+    aws s3 cp "/tmp/garage.$idx" "s3://eprouvette/&+-é\"/garage.$idx.aws"
     aws s3 ls s3://eprouvette
     aws s3 cp "s3://eprouvette/&+-é\"/garage.$idx.aws" "/tmp/garage.$idx.dl"
-    diff /tmp/garage.$idx.rnd /tmp/garage.$idx.dl
+    diff /tmp/garage.$idx /tmp/garage.$idx.dl
     rm /tmp/garage.$idx.dl
     aws s3 rm "s3://eprouvette/&+-é\"/garage.$idx.aws"
   done
@@ -51,11 +58,11 @@ if [ -z "$SKIP_S3CMD" ]; then
   echo "🛠️ Testing with s3cmd"
   source ${SCRIPT_FOLDER}/dev-env-s3cmd.sh
   s3cmd ls
-  for idx in $(seq 1 3); do
-    s3cmd put "/tmp/garage.$idx.rnd" "s3://eprouvette/&+-é\"/garage.$idx.s3cmd"
+  for idx in {1..3}.{rnd,b64}; do
+    s3cmd put "/tmp/garage.$idx" "s3://eprouvette/&+-é\"/garage.$idx.s3cmd"
     s3cmd ls s3://eprouvette
     s3cmd get "s3://eprouvette/&+-é\"/garage.$idx.s3cmd" "/tmp/garage.$idx.dl"
-    diff /tmp/garage.$idx.rnd /tmp/garage.$idx.dl
+    diff /tmp/garage.$idx /tmp/garage.$idx.dl
     rm /tmp/garage.$idx.dl
     s3cmd rm "s3://eprouvette/&+-é\"/garage.$idx.s3cmd"
   done
@@ -66,11 +73,11 @@ if [ -z "$SKIP_MC" ]; then
   echo "🛠️ Testing with mc (minio client)"
   source ${SCRIPT_FOLDER}/dev-env-mc.sh
   mc ls garage/
-  for idx in $(seq 1 3); do
-    mc cp "/tmp/garage.$idx.rnd" "garage/eprouvette/&+-é\"/garage.$idx.mc"
+  for idx in {1..3}.{rnd,b64}; do
+    mc cp "/tmp/garage.$idx" "garage/eprouvette/&+-é\"/garage.$idx.mc"
     mc ls garage/eprouvette
     mc cp "garage/eprouvette/&+-é\"/garage.$idx.mc" "/tmp/garage.$idx.dl"
-    diff /tmp/garage.$idx.rnd /tmp/garage.$idx.dl
+    diff /tmp/garage.$idx /tmp/garage.$idx.dl
     rm /tmp/garage.$idx.dl
     mc rm "garage/eprouvette/&+-é\"/garage.$idx.mc"
   done
@@ -81,13 +88,13 @@ if [ -z "$SKIP_RCLONE" ]; then
   echo "🛠️ Testing with rclone"
   source ${SCRIPT_FOLDER}/dev-env-rclone.sh
   rclone lsd garage:
-  for idx in $(seq 1 3); do
-    cp /tmp/garage.$idx.rnd /tmp/garage.$idx.dl
+  for idx in {1..3}.{rnd,b64}; do
+    cp /tmp/garage.$idx /tmp/garage.$idx.dl
     rclone copy "/tmp/garage.$idx.dl" "garage:eprouvette/&+-é\"/"
     rm /tmp/garage.$idx.dl
     rclone ls garage:eprouvette
     rclone copy "garage:eprouvette/&+-é\"/garage.$idx.dl" "/tmp/"
-    diff /tmp/garage.$idx.rnd /tmp/garage.$idx.dl
+    diff /tmp/garage.$idx /tmp/garage.$idx.dl
     rm /tmp/garage.$idx.dl
     rclone delete "garage:eprouvette/&+-é\"/garage.$idx.dl"
   done
@@ -99,27 +106,249 @@ if [ -z "$SKIP_DUCK" ]; then
   source ${SCRIPT_FOLDER}/dev-env-duck.sh
   duck --list garage:/
   duck --mkdir "garage:/eprouvette/duck"
-  for idx in $(seq 1 3); do
-    duck --verbose --upload "garage:/eprouvette/duck/" "/tmp/garage.$idx.rnd" 
+  for idx in {1..3}.{rnd,b64}; do
+    duck --verbose --upload "garage:/eprouvette/duck/" "/tmp/garage.$idx"
     duck --list garage:/eprouvette/duck/
-    duck --download "garage:/eprouvette/duck/garage.$idx.rnd" "/tmp/garage.$idx.dl"
-    diff /tmp/garage.$idx.rnd /tmp/garage.$idx.dl
+    duck --download "garage:/eprouvette/duck/garage.$idx" "/tmp/garage.$idx.dl"
+    diff /tmp/garage.$idx /tmp/garage.$idx.dl
     rm /tmp/garage.$idx.dl
     duck --delete "garage:/eprouvette/duck/garage.$idx.dk"
   done
 fi
 
-rm /tmp/garage.{1,2,3}.rnd
+# Advanced testing via S3API
+if [ -z "$SKIP_AWS" ]; then
+  echo "🔌 Test S3API"
+
+  echo "Test Objects"
+  aws s3api put-object --bucket eprouvette --key a
+  aws s3api put-object --bucket eprouvette --key a/a
+  aws s3api put-object --bucket eprouvette --key a/b
+  aws s3api put-object --bucket eprouvette --key a/c
+  aws s3api put-object --bucket eprouvette --key a/d/a
+  aws s3api put-object --bucket eprouvette --key a/é
+  aws s3api put-object --bucket eprouvette --key b
+  aws s3api put-object --bucket eprouvette --key c
+
+
+  aws s3api list-objects-v2 --bucket eprouvette >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 8 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects-v2 --bucket eprouvette --page-size 0 >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 8 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects-v2 --bucket eprouvette --page-size 999999999999999 >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 8 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects-v2 --bucket eprouvette --page-size 1 >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 8 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects-v2 --bucket eprouvette --delimiter '/' >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 3 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-objects-v2 --bucket eprouvette --delimiter '/' --page-size 1 >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 3 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-objects-v2 --bucket eprouvette --prefix 'a/' >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 5 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects-v2 --bucket eprouvette --prefix 'a/' --delimiter '/' >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 4 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-objects-v2 --bucket eprouvette --prefix 'a/' --page-size 1 >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 5 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects-v2 --bucket eprouvette --prefix 'a/' --delimiter '/' --page-size 1 >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 4 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-objects-v2 --bucket eprouvette --start-after 'Z' >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 8 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects-v2 --bucket eprouvette --start-after 'c' >$CMDOUT
+  ! [ -s $CMDOUT ]
+
+
+  aws s3api list-objects --bucket eprouvette >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 8 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects --bucket eprouvette --page-size 1 >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 8 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects --bucket eprouvette --delimiter '/' >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 3 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  # @FIXME it does not work as expected but might be a limitation of aws s3api
+  # The problem is the conjunction of a delimiter + pagination + v1 of listobjects
+  #aws s3api list-objects --bucket eprouvette --delimiter '/' --page-size 1 >$CMDOUT
+  #[ $(jq '.Contents | length' $CMDOUT) == 3 ]
+  #[ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-objects --bucket eprouvette --prefix 'a/' >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 5 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects --bucket eprouvette --prefix 'a/' --delimiter '/' >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 4 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-objects --bucket eprouvette --prefix 'a/' --page-size 1 >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 5 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  # @FIXME idem
+  #aws s3api list-objects --bucket eprouvette --prefix 'a/' --delimiter '/' --page-size 1 >$CMDOUT
+  #[ $(jq '.Contents | length' $CMDOUT) == 4 ]
+  #[ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-objects --bucket eprouvette --starting-token 'Z' >$CMDOUT
+  [ $(jq '.Contents | length' $CMDOUT) == 8 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-objects --bucket eprouvette --starting-token 'c' >$CMDOUT
+  ! [ -s $CMDOUT ]
+
+  aws s3api list-objects-v2 --bucket eprouvette | \
+    jq -c '. | {Objects: [.Contents[] | {Key: .Key}], Quiet: true}' | \
+    aws s3api delete-objects --bucket eprouvette --delete file:///dev/stdin
+
+
+  echo "Test Multipart Upload"
+  aws s3api create-multipart-upload --bucket eprouvette --key a
+  aws s3api create-multipart-upload --bucket eprouvette --key a
+  aws s3api create-multipart-upload --bucket eprouvette --key c
+  aws s3api create-multipart-upload --bucket eprouvette --key c/a
+  aws s3api create-multipart-upload --bucket eprouvette --key c/b
+
+  aws s3api list-multipart-uploads --bucket eprouvette >$CMDOUT
+  [ $(jq '.Uploads | length' $CMDOUT) == 5 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-multipart-uploads --bucket eprouvette --page-size 1 >$CMDOUT
+  [ $(jq '.Uploads | length' $CMDOUT) == 5 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-multipart-uploads --bucket eprouvette --delimiter '/' >$CMDOUT
+  [ $(jq '.Uploads | length' $CMDOUT) == 3 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-multipart-uploads --bucket eprouvette --delimiter '/' --page-size 1 >$CMDOUT
+  [ $(jq '.Uploads | length' $CMDOUT) == 3 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-multipart-uploads --bucket eprouvette --prefix 'c' >$CMDOUT
+  [ $(jq '.Uploads | length' $CMDOUT) == 3 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-multipart-uploads --bucket eprouvette --prefix 'c' --page-size 1 >$CMDOUT
+  [ $(jq '.Uploads | length' $CMDOUT) == 3 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-multipart-uploads --bucket eprouvette --prefix 'c' --delimiter '/' >$CMDOUT
+  [ $(jq '.Uploads | length' $CMDOUT) == 1 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-multipart-uploads --bucket eprouvette --prefix 'c' --delimiter '/' --page-size 1 >$CMDOUT
+  [ $(jq '.Uploads | length' $CMDOUT) == 1 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 1 ]
+  aws s3api list-multipart-uploads --bucket eprouvette --starting-token 'ZZZZZ' >$CMDOUT
+  [ $(jq '.Uploads | length' $CMDOUT) == 5 ]
+  [ $(jq '.CommonPrefixes | length' $CMDOUT) == 0 ]
+  aws s3api list-multipart-uploads --bucket eprouvette --starting-token 'd' >$CMDOUT
+  ! [ -s $CMDOUT ]
+
+  aws s3api list-multipart-uploads --bucket eprouvette | \
+    jq -r '.Uploads[] | "\(.Key) \(.UploadId)"' | \
+    while read r; do 
+      key=$(echo $r|cut -d' ' -f 1); 
+      uid=$(echo $r|cut -d' ' -f 2); 
+      aws s3api abort-multipart-upload --bucket eprouvette --key $key --upload-id $uid;
+      echo "Deleted ${key}:${uid}"
+    done
+
+  # Test for UploadPartCopy
+  aws s3 cp "/tmp/garage.3.rnd" "s3://eprouvette/copy_part_source"
+  UPLOAD_ID=$(aws s3api create-multipart-upload --bucket eprouvette --key test_multipart | jq -r .UploadId)
+  PART1=$(aws s3api upload-part \
+    --bucket eprouvette --key test_multipart \
+    --upload-id $UPLOAD_ID --part-number 1 \
+    --body /tmp/garage.2.rnd | jq .ETag)
+  PART2=$(aws s3api upload-part-copy \
+    --bucket eprouvette --key test_multipart \
+    --upload-id $UPLOAD_ID --part-number 2 \
+    --copy-source "/eprouvette/copy_part_source" \
+    --copy-source-range "bytes=500-5000500" \
+    | jq .CopyPartResult.ETag)
+  PART3=$(aws s3api upload-part \
+    --bucket eprouvette --key test_multipart \
+    --upload-id $UPLOAD_ID --part-number 3 \
+    --body /tmp/garage.3.rnd | jq .ETag)
+  cat >/tmp/garage.multipart_struct <<EOF
+{
+  "Parts": [
+    {
+      "ETag": $PART1,
+      "PartNumber": 1
+    },
+    {
+      "ETag": $PART2,
+      "PartNumber": 2
+    },
+    {
+      "ETag": $PART3,
+      "PartNumber": 3
+    }
+  ]
+}
+EOF
+  aws s3api complete-multipart-upload \
+    --bucket eprouvette --key test_multipart --upload-id $UPLOAD_ID \
+    --multipart-upload file:///tmp/garage.multipart_struct
+
+  aws s3 cp "s3://eprouvette/test_multipart" /tmp/garage.test_multipart
+  cat /tmp/garage.2.rnd <(tail -c +501 /tmp/garage.3.rnd | head -c 5000001) /tmp/garage.3.rnd > /tmp/garage.test_multipart_reference
+  diff /tmp/garage.test_multipart /tmp/garage.test_multipart_reference >/tmp/garage.test_multipart_diff 2>&1
+
+  aws s3 rm "s3://eprouvette/copy_part_source"
+  aws s3 rm "s3://eprouvette/test_multipart"
+
+  rm /tmp/garage.multipart_struct
+  rm /tmp/garage.test_multipart
+  rm /tmp/garage.test_multipart_reference
+  rm /tmp/garage.test_multipart_diff
+
+  echo "Test CORS endpoints"
+  # @FIXME remove bucket allow if/when testing on s3 endpoint
+  garage -c /tmp/config.1.toml bucket website --allow eprouvette
+  aws s3api put-object --bucket eprouvette --key index.html
+  CORS='{"CORSRules":[{"AllowedHeaders":["*"],"AllowedMethods":["GET","PUT"],"AllowedOrigins":["*"]}]}'
+  aws s3api put-bucket-cors --bucket eprouvette --cors-configuration $CORS
+  [ `aws s3api get-bucket-cors --bucket eprouvette | jq -c` == $CORS ]
+
+  # @FIXME should we really return these CORS on the WEB endpoint and not on the S3 endpoint?
+  curl -s -i -H 'Origin: http://example.com' http://eprouvette.web.garage.localhost:3921 | grep access-control-allow-origin
+  curl -s -i -X OPTIONS -H 'Access-Control-Request-Method: PUT' -H 'Origin: http://example.com' http://eprouvette.web.garage.localhost:3921|grep access-control-allow-methods
+  curl -s -i -X OPTIONS -H 'Access-Control-Request-Method: DELETE' -H 'Origin: http://example.com' http://eprouvette.web.garage.localhost:3921 |grep '403 Forbidden'
+
+  aws s3api delete-bucket-cors --bucket eprouvette 
+  ! [ -s `aws s3api get-bucket-cors --bucket eprouvette` ]
+  curl -s -i -X OPTIONS -H 'Access-Control-Request-Method: PUT' -H 'Origin: http://example.com' http://eprouvette.web.garage.localhost:3921|grep '403 Forbidden'
+  aws s3api delete-object --bucket eprouvette --key index.html
+  garage -c /tmp/config.1.toml bucket website --deny eprouvette
+fi
+
+rm /tmp/garage.{1..3}.{rnd,b64}
+
+if [ -z "$SKIP_AWS" ]; then
+  echo "🪣 Test bucket logic "
+  AWS_ACCESS_KEY_ID=`cat /tmp/garage.s3 |cut -d' ' -f1`
+  [ $(aws s3 ls | wc -l) == 1 ]
+  garage -c /tmp/config.1.toml bucket create seau
+  garage -c /tmp/config.1.toml bucket allow --read seau --key $AWS_ACCESS_KEY_ID
+  [ $(aws s3 ls | wc -l) == 2 ]
+  garage -c /tmp/config.1.toml bucket deny --read seau --key $AWS_ACCESS_KEY_ID
+  [ $(aws s3 ls | wc -l) == 1 ]
+  garage -c /tmp/config.1.toml bucket allow --read seau --key $AWS_ACCESS_KEY_ID
+  [ $(aws s3 ls | wc -l) == 2 ]
+  garage -c /tmp/config.1.toml bucket delete --yes seau
+  [ $(aws s3 ls | wc -l) == 1 ]
+fi
 
 if [ -z "$SKIP_AWS" ]; then
   echo "🧪 Website Testing"
   echo "<h1>hello world</h1>" > /tmp/garage-index.html
   aws s3 cp /tmp/garage-index.html s3://eprouvette/index.html
-  [ `curl -s -o /dev/null -w "%{http_code}" --header "Host: eprouvette.garage.tld"  http://127.0.0.1:3923/ ` == 404 ]
-  garage bucket website --allow eprouvette
-  [ `curl -s -o /dev/null -w "%{http_code}" --header "Host: eprouvette.garage.tld"  http://127.0.0.1:3923/ ` == 200 ]
-  garage bucket website --deny eprouvette
-  [ `curl -s -o /dev/null -w "%{http_code}" --header "Host: eprouvette.garage.tld"  http://127.0.0.1:3923/ ` == 404 ]
+  [ `curl -s -o /dev/null -w "%{http_code}" --header "Host: eprouvette.web.garage.localhost"  http://127.0.0.1:3921/ ` == 404 ]
+  garage -c /tmp/config.1.toml bucket website --allow eprouvette
+  [ `curl -s -o /dev/null -w "%{http_code}" --header "Host: eprouvette.web.garage.localhost"  http://127.0.0.1:3921/ ` == 200 ]
+  garage -c /tmp/config.1.toml bucket website --deny eprouvette
+  [ `curl -s -o /dev/null -w "%{http_code}" --header "Host: eprouvette.web.garage.localhost"  http://127.0.0.1:3921/ ` == 404 ]
   aws s3 rm s3://eprouvette/index.html
   rm /tmp/garage-index.html
 fi
@@ -127,8 +356,9 @@ fi
 echo "🏁 Teardown"
 AWS_ACCESS_KEY_ID=`cat /tmp/garage.s3 |cut -d' ' -f1`
 AWS_SECRET_ACCESS_KEY=`cat /tmp/garage.s3 |cut -d' ' -f2`
-garage bucket deny --read --write eprouvette --key $AWS_ACCESS_KEY_ID
-garage bucket delete --yes eprouvette
-garage key delete --yes $AWS_ACCESS_KEY_ID
+garage -c /tmp/config.1.toml bucket deny --read --write eprouvette --key $AWS_ACCESS_KEY_ID
+garage -c /tmp/config.1.toml bucket delete --yes eprouvette
+garage -c /tmp/config.1.toml key delete --yes $AWS_ACCESS_KEY_ID
+exec 3>&-
 
 echo "✅ Success"

shell.nix

@@ -0,0 +1,94 @@
+{
+  system ? builtins.currentSystem,
+  rust ? true,
+  integration ? true,
+  release ? true,
+}:
+
+with import ./nix/common.nix;
+
+let
+  pkgs = import pkgsSrc {
+    inherit system;
+    overlays = [ cargo2nixOverlay ];
+  };
+  kaniko = (import ./nix/kaniko.nix) pkgs;
+
+in
+
+pkgs.mkShell {
+  shellHook = ''
+function to_s3 {
+  aws \
+      --endpoint-url https://garage.deuxfleurs.fr \
+      --region garage \
+    s3 cp \
+      ./result/bin/garage \
+      s3://garagehq.deuxfleurs.fr/_releases/''${DRONE_TAG:-$DRONE_COMMIT}/''${TARGET}/garage
+}
+
+function to_docker {
+  executor  \
+    --force \
+    --customPlatform="''${DOCKER_PLATFORM}" \
+    --destination "''${CONTAINER_NAME}:''${CONTAINER_TAG}" \
+    --context dir://`pwd` \
+    --verbosity=debug
+}
+
+function refresh_index {
+  aws \
+      --endpoint-url https://garage.deuxfleurs.fr \
+      --region garage \
+    s3 ls \
+      --recursive \
+      s3://garagehq.deuxfleurs.fr/_releases/ \
+  > aws-list.txt
+
+  nix-build nix/build_index.nix
+
+  aws \
+      --endpoint-url https://garage.deuxfleurs.fr \
+      --region garage \
+    s3 cp \
+      --content-type "text/html" \
+      result \
+      s3://garagehq.deuxfleurs.fr/_releases.html
+}
+
+function refresh_toolchain {
+  nix copy \
+    --to 's3://nix?endpoint=garage.deuxfleurs.fr&region=garage&secret-key=/etc/nix/signing-key.sec' \
+    $(nix-store -qR \
+      $(nix-build --quiet --no-build-output --no-out-link nix/toolchain.nix))
+}
+  '';
+
+  nativeBuildInputs = 
+   (if rust then [
+     pkgs.rustPlatform.rust.rustc
+     pkgs.rustPlatform.rust.cargo
+     pkgs.clippy
+     pkgs.rustfmt
+     /*(pkgs.callPackage cargo2nix {}).package*/
+    ] else [])
+   ++
+   (if integration then [ 
+     pkgs.s3cmd
+     pkgs.awscli2
+     pkgs.minio-client
+     pkgs.rclone
+     pkgs.socat
+     pkgs.psmisc
+     pkgs.which
+     pkgs.openssl
+     pkgs.curl
+     pkgs.jq
+    ] else [])
+   ++
+   (if release then [ 
+     pkgs.awscli2
+     kaniko 
+    ] else [])
+   ;
+}

src/api/Cargo.toml

@@ -1,11 +1,12 @@
 [package]
 name = "garage_api"
-version = "0.3.0"
+version = "0.6.0"
 authors = ["Alex Auvolat <alex@adnab.me>"]
 edition = "2018"
 license = "AGPL-3.0"
 description = "S3 API server crate for the Garage object store"
 repository = "https://git.deuxfleurs.fr/Deuxfleurs/garage"
+readme = "../../README.md"
 
 [lib]
 path = "lib.rs"
@@ -13,9 +14,9 @@ path = "lib.rs"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-garage_model = { version = "0.3.0", path = "../model" }
-garage_table = { version = "0.3.0", path = "../table" }
-garage_util = { version = "0.3.0", path = "../util" }
+garage_model = { version = "0.6.0", path = "../model" }
+garage_table = { version = "0.6.0", path = "../table" }
+garage_util = { version = "0.6.0", path = "../util" }
 
 base64 = "0.13"
 bytes = "1.0"
@@ -24,6 +25,7 @@ crypto-mac = "0.10"
 err-derive = "0.3"
 hex = "0.4"
 hmac = "0.10"
+idna = "0.2"
 log = "0.4"
 md-5 = "0.9"
 sha2 = "0.9"
@@ -35,9 +37,10 @@ tokio = { version = "1.0", default-features = false, features = ["rt", "rt-multi
 http = "0.2"
 httpdate = "0.3"
 http-range = "0.1"
-hyper = "0.14"
+hyper = { version = "0.14", features = ["server", "http1", "runtime", "tcp", "stream"] }
 percent-encoding = "2.1.0"
 roxmltree = "0.14"
 serde = { version = "1.0", features = ["derive"] }
+serde_bytes = "0.11"
 quick-xml = { version = "0.21", features = [ "serialize" ] }
 url = "2.1"

src/api/api_server.rs

@@ -1,25 +1,32 @@
-use std::collections::HashMap;
+use std::cmp::{max, min};
 use std::net::SocketAddr;
 use std::sync::Arc;
 
 use futures::future::Future;
+use hyper::header;
 use hyper::server::conn::AddrStream;
 use hyper::service::{make_service_fn, service_fn};
-use hyper::{Body, Method, Request, Response, Server};
+use hyper::{Body, Request, Response, Server};
 
+use garage_util::data::*;
 use garage_util::error::Error as GarageError;
 
 use garage_model::garage::Garage;
+use garage_model::key_table::Key;
 
 use crate::error::*;
 use crate::signature::check_signature;
 
+use crate::helpers::*;
 use crate::s3_bucket::*;
 use crate::s3_copy::*;
+use crate::s3_cors::*;
 use crate::s3_delete::*;
 use crate::s3_get::*;
 use crate::s3_list::*;
 use crate::s3_put::*;
+use crate::s3_router::{Authorization, Endpoint};
+use crate::s3_website::*;
 
 /// Run the S3 API server
 pub async fn run_api_server(
@@ -39,7 +46,7 @@ pub async fn run_api_server(
 		}
 	});
 
-	let server = Server::bind(&addr).serve(service);
+	let server = Server::bind(addr).serve(service);
 
 	let graceful = server.with_graceful_shutdown(shutdown_signal);
 	info!("API server listening on http://{}", addr);
@@ -63,10 +70,15 @@ async fn handler(
 		}
 		Err(e) => {
 			let body: Body = Body::from(e.aws_xml(&garage.config.s3_api.s3_region, uri.path()));
-			let http_error = Response::builder()
+			let mut http_error_builder = Response::builder()
 				.status(e.http_status_code())
-				.header("Content-Type", "application/xml")
-				.body(body)?;
+				.header("Content-Type", "application/xml");
+
+			if let Some(header_map) = http_error_builder.headers_mut() {
+				e.add_headers(header_map)
+			}
+
+			let http_error = http_error_builder.body(body)?;
 
 			if e.http_status_code().is_server_error() {
 				warn!("Response: error {}, {}", e.http_status_code(), e);
@@ -79,182 +91,288 @@ async fn handler(
 }
 
 async fn handler_inner(garage: Arc<Garage>, req: Request<Body>) -> Result<Response<Body>, Error> {
-	let path = req.uri().path().to_string();
-	let path = percent_encoding::percent_decode_str(&path).decode_utf8()?;
 	let (api_key, content_sha256) = check_signature(&garage, &req).await?;
-	if path == "/" {
-		return handle_list_buckets(&api_key);
+
+	let authority = req
+		.headers()
+		.get(header::HOST)
+		.ok_or_else(|| Error::BadRequest("HOST header required".to_owned()))?
+		.to_str()?;
+
+	let host = authority_to_host(authority)?;
+
+	let bucket = garage
+		.config
+		.s3_api
+		.root_domain
+		.as_ref()
+		.and_then(|root_domain| host_to_bucket(&host, root_domain));
+
+	let endpoint = Endpoint::from_request(&req, bucket.map(ToOwned::to_owned))?;
+	debug!("Endpoint: {:?}", endpoint);
+
+	// Special code path for CreateBucket API endpoint
+	if let Endpoint::CreateBucket { bucket } = endpoint {
+		return handle_create_bucket(&garage, req, content_sha256, api_key, bucket).await;
 	}
 
-	let (bucket, key) = parse_bucket_key(&path)?;
-	let allowed = match req.method() {
-		&Method::HEAD | &Method::GET => api_key.allow_read(&bucket),
-		_ => api_key.allow_write(&bucket),
+	let bucket_name = match endpoint.get_bucket() {
+		None => return handle_request_without_bucket(garage, req, api_key, endpoint).await,
+		Some(bucket) => bucket.to_string(),
 	};
+
+	let bucket_id = resolve_bucket(&garage, &bucket_name, &api_key).await?;
+
+	let allowed = match endpoint.authorization_type() {
+		Authorization::Read(_) => api_key.allow_read(&bucket_id),
+		Authorization::Write(_) => api_key.allow_write(&bucket_id),
+		Authorization::Owner(_) => api_key.allow_owner(&bucket_id),
+		_ => unreachable!(),
+	};
+
 	if !allowed {
 		return Err(Error::Forbidden(
 			"Operation is not allowed for this key.".to_string(),
 		));
 	}
 
-	let mut params = HashMap::new();
-	if let Some(query) = req.uri().query() {
-		let query_pairs = url::form_urlencoded::parse(query.as_bytes());
-		for (key, val) in query_pairs {
-			params.insert(key.to_lowercase(), val.to_string());
+	match endpoint {
+		Endpoint::HeadObject { key, .. } => handle_head(garage, &req, bucket_id, &key).await,
+		Endpoint::GetObject { key, .. } => handle_get(garage, &req, bucket_id, &key).await,
+		Endpoint::UploadPart {
+			key,
+			part_number,
+			upload_id,
+			..
+		} => {
+			handle_put_part(
+				garage,
+				req,
+				bucket_id,
+				&key,
+				part_number,
+				&upload_id,
+				content_sha256,
+			)
+			.await
 		}
-	}
-
-	if let Some(key) = key {
-		match *req.method() {
-			Method::HEAD => {
-				// HeadObject query
-				Ok(handle_head(garage, &req, &bucket, &key).await?)
-			}
-			Method::GET => {
-				// GetObject query
-				Ok(handle_get(garage, &req, &bucket, &key).await?)
-			}
-			Method::PUT => {
-				if params.contains_key(&"partnumber".to_string())
-					&& params.contains_key(&"uploadid".to_string())
-				{
-					// UploadPart query
-					let part_number = params.get("partnumber").unwrap();
-					let upload_id = params.get("uploadid").unwrap();
-					Ok(handle_put_part(
-						garage,
-						req,
-						&bucket,
-						&key,
-						part_number,
-						upload_id,
-						content_sha256,
-					)
-					.await?)
-				} else if req.headers().contains_key("x-amz-copy-source") {
-					// CopyObject query
-					let copy_source = req.headers().get("x-amz-copy-source").unwrap().to_str()?;
-					let copy_source =
-						percent_encoding::percent_decode_str(&copy_source).decode_utf8()?;
-					let (source_bucket, source_key) = parse_bucket_key(&copy_source)?;
-					if !api_key.allow_read(&source_bucket) {
-						return Err(Error::Forbidden(format!(
-							"Reading from bucket {} not allowed for this key",
-							source_bucket
-						)));
-					}
-					let source_key = source_key.ok_or_bad_request("No source key specified")?;
-					Ok(
-						handle_copy(garage, &req, &bucket, &key, &source_bucket, &source_key)
-							.await?,
-					)
-				} else {
-					// PutObject query
-					Ok(handle_put(garage, req, &bucket, &key, content_sha256).await?)
-				}
-			}
-			Method::DELETE => {
-				if params.contains_key(&"uploadid".to_string()) {
-					// AbortMultipartUpload query
-					let upload_id = params.get("uploadid").unwrap();
-					Ok(handle_abort_multipart_upload(garage, &bucket, &key, upload_id).await?)
-				} else {
-					// DeleteObject query
-					Ok(handle_delete(garage, &bucket, &key).await?)
-				}
-			}
-			Method::POST => {
-				if params.contains_key(&"uploads".to_string()) {
-					// CreateMultipartUpload call
-					Ok(handle_create_multipart_upload(garage, &req, &bucket, &key).await?)
-				} else if params.contains_key(&"uploadid".to_string()) {
-					// CompleteMultipartUpload call
-					let upload_id = params.get("uploadid").unwrap();
-					Ok(handle_complete_multipart_upload(
-						garage,
-						req,
-						&bucket,
-						&key,
-						upload_id,
-						content_sha256,
-					)
-					.await?)
-				} else {
-					Err(Error::BadRequest(
-						"Not a CreateMultipartUpload call, what is it?".to_string(),
-					))
-				}
-			}
-			_ => Err(Error::BadRequest("Invalid method".to_string())),
+		Endpoint::CopyObject { key, .. } => {
+			handle_copy(garage, &api_key, &req, bucket_id, &key).await
 		}
-	} else {
-		match *req.method() {
-			Method::PUT => {
-				// CreateBucket
-				// If we're here, the bucket already exists, so just answer ok
-				debug!(
-					"Body: {}",
-					std::str::from_utf8(&hyper::body::to_bytes(req.into_body()).await?)
-						.unwrap_or("<invalid utf8>")
-				);
-				let empty_body: Body = Body::from(vec![]);
-				let response = Response::builder()
-					.header("Location", format!("/{}", bucket))
-					.body(empty_body)
-					.unwrap();
-				Ok(response)
-			}
-			Method::HEAD => {
-				// HeadBucket
-				let empty_body: Body = Body::from(vec![]);
-				let response = Response::builder().body(empty_body).unwrap();
-				Ok(response)
-			}
-			Method::DELETE => {
-				// DeleteBucket query
-				Err(Error::Forbidden(
-					"Cannot delete buckets using S3 api, please talk to Garage directly".into(),
-				))
-			}
-			Method::GET => {
-				if params.contains_key("location") {
-					// GetBucketLocation call
-					Ok(handle_get_bucket_location(garage)?)
-				} else if params.contains_key("versioning") {
-					// GetBucketVersioning
-					Ok(handle_get_bucket_versioning()?)
-				} else {
-					// ListObjects or ListObjectsV2 query
-					let q = parse_list_objects_query(bucket, &params)?;
-					Ok(handle_list(garage, &q).await?)
-				}
-			}
-			Method::POST => {
-				if params.contains_key(&"delete".to_string()) {
-					// DeleteObjects
-					Ok(handle_delete_objects(garage, bucket, req, content_sha256).await?)
-				} else {
-					debug!(
-						"Body: {}",
-						std::str::from_utf8(&hyper::body::to_bytes(req.into_body()).await?)
-							.unwrap_or("<invalid utf8>")
-					);
-					Err(Error::BadRequest("Unsupported call".to_string()))
-				}
-			}
-			_ => Err(Error::BadRequest("Invalid method".to_string())),
+		Endpoint::UploadPartCopy {
+			key,
+			part_number,
+			upload_id,
+			..
+		} => {
+			handle_upload_part_copy(
+				garage,
+				&api_key,
+				&req,
+				bucket_id,
+				&key,
+				part_number,
+				&upload_id,
+			)
+			.await
 		}
+		Endpoint::PutObject { key, .. } => {
+			handle_put(garage, req, bucket_id, &key, content_sha256).await
+		}
+		Endpoint::AbortMultipartUpload { key, upload_id, .. } => {
+			handle_abort_multipart_upload(garage, bucket_id, &key, &upload_id).await
+		}
+		Endpoint::DeleteObject { key, .. } => handle_delete(garage, bucket_id, &key).await,
+		Endpoint::CreateMultipartUpload { bucket, key } => {
+			handle_create_multipart_upload(garage, &req, &bucket, bucket_id, &key).await
+		}
+		Endpoint::CompleteMultipartUpload {
+			bucket,
+			key,
+			upload_id,
+		} => {
+			handle_complete_multipart_upload(
+				garage,
+				req,
+				&bucket,
+				bucket_id,
+				&key,
+				&upload_id,
+				content_sha256,
+			)
+			.await
+		}
+		Endpoint::CreateBucket { .. } => unreachable!(),
+		Endpoint::HeadBucket { .. } => {
+			let empty_body: Body = Body::from(vec![]);
+			let response = Response::builder().body(empty_body).unwrap();
+			Ok(response)
+		}
+		Endpoint::DeleteBucket { .. } => {
+			handle_delete_bucket(&garage, bucket_id, bucket_name, api_key).await
+		}
+		Endpoint::GetBucketLocation { .. } => handle_get_bucket_location(garage),
+		Endpoint::GetBucketVersioning { .. } => handle_get_bucket_versioning(),
+		Endpoint::ListObjects {
+			bucket,
+			delimiter,
+			encoding_type,
+			marker,
+			max_keys,
+			prefix,
+		} => {
+			handle_list(
+				garage,
+				&ListObjectsQuery {
+					common: ListQueryCommon {
+						bucket_name: bucket,
+						bucket_id,
+						delimiter: delimiter.map(|d| d.to_string()),
+						page_size: max_keys.map(|p| min(1000, max(1, p))).unwrap_or(1000),
+						prefix: prefix.unwrap_or_default(),
+						urlencode_resp: encoding_type.map(|e| e == "url").unwrap_or(false),
+					},
+					is_v2: false,
+					marker,
+					continuation_token: None,
+					start_after: None,
+				},
+			)
+			.await
+		}
+		Endpoint::ListObjectsV2 {
+			bucket,
+			delimiter,
+			encoding_type,
+			max_keys,
+			prefix,
+			continuation_token,
+			start_after,
+			list_type,
+			..
+		} => {
+			if list_type == "2" {
+				handle_list(
+					garage,
+					&ListObjectsQuery {
+						common: ListQueryCommon {
+							bucket_name: bucket,
+							bucket_id,
+							delimiter: delimiter.map(|d| d.to_string()),
+							page_size: max_keys.map(|p| min(1000, max(1, p))).unwrap_or(1000),
+							urlencode_resp: encoding_type.map(|e| e == "url").unwrap_or(false),
+							prefix: prefix.unwrap_or_default(),
+						},
+						is_v2: true,
+						marker: None,
+						continuation_token,
+						start_after,
+					},
+				)
+				.await
+			} else {
+				Err(Error::BadRequest(format!(
+					"Invalid endpoint: list-type={}",
+					list_type
+				)))
+			}
+		}
+		Endpoint::ListMultipartUploads {
+			bucket,
+			delimiter,
+			encoding_type,
+			key_marker,
+			max_uploads,
+			prefix,
+			upload_id_marker,
+		} => {
+			handle_list_multipart_upload(
+				garage,
+				&ListMultipartUploadsQuery {
+					common: ListQueryCommon {
+						bucket_name: bucket,
+						bucket_id,
+						delimiter: delimiter.map(|d| d.to_string()),
+						page_size: max_uploads.map(|p| min(1000, max(1, p))).unwrap_or(1000),
+						prefix: prefix.unwrap_or_default(),
+						urlencode_resp: encoding_type.map(|e| e == "url").unwrap_or(false),
+					},
+					key_marker,
+					upload_id_marker,
+				},
+			)
+			.await
+		}
+		Endpoint::DeleteObjects { .. } => {
+			handle_delete_objects(garage, bucket_id, req, content_sha256).await
+		}
+		Endpoint::GetBucketWebsite { .. } => handle_get_website(garage, bucket_id).await,
+		Endpoint::PutBucketWebsite { .. } => {
+			handle_put_website(garage, bucket_id, req, content_sha256).await
+		}
+		Endpoint::DeleteBucketWebsite { .. } => handle_delete_website(garage, bucket_id).await,
+		Endpoint::GetBucketCors { .. } => handle_get_cors(garage, bucket_id).await,
+		Endpoint::PutBucketCors { .. } => {
+			handle_put_cors(garage, bucket_id, req, content_sha256).await
+		}
+		Endpoint::DeleteBucketCors { .. } => handle_delete_cors(garage, bucket_id).await,
+		endpoint => Err(Error::NotImplemented(endpoint.name().to_owned())),
 	}
 }
 
-/// Extract the bucket name and the key name from an HTTP path
+async fn handle_request_without_bucket(
+	garage: Arc<Garage>,
+	_req: Request<Body>,
+	api_key: Key,
+	endpoint: Endpoint,
+) -> Result<Response<Body>, Error> {
+	match endpoint {
+		Endpoint::ListBuckets => handle_list_buckets(&garage, &api_key).await,
+		endpoint => Err(Error::NotImplemented(endpoint.name().to_owned())),
+	}
+}
+
+#[allow(clippy::ptr_arg)]
+pub async fn resolve_bucket(
+	garage: &Garage,
+	bucket_name: &String,
+	api_key: &Key,
+) -> Result<Uuid, Error> {
+	let api_key_params = api_key
+		.state
+		.as_option()
+		.ok_or_internal_error("Key should not be deleted at this point")?;
+
+	if let Some(Some(bucket_id)) = api_key_params.local_aliases.get(bucket_name) {
+		Ok(*bucket_id)
+	} else {
+		Ok(garage
+			.bucket_helper()
+			.resolve_global_bucket_name(bucket_name)
+			.await?
+			.ok_or(Error::NoSuchBucket)?)
+	}
+}
+
+/// Extract the bucket name and the key name from an HTTP path and possibly a bucket provided in
+/// the host header of the request
 ///
 /// S3 internally manages only buckets and keys. This function splits
 /// an HTTP path to get the corresponding bucket name and key.
-fn parse_bucket_key(path: &str) -> Result<(&str, Option<&str>), Error> {
+pub fn parse_bucket_key<'a>(
+	path: &'a str,
+	host_bucket: Option<&'a str>,
+) -> Result<(&'a str, Option<&'a str>), Error> {
 	let path = path.trim_start_matches('/');
 
+	if let Some(bucket) = host_bucket {
+		if !path.is_empty() {
+			return Ok((bucket, Some(path)));
+		} else {
+			return Ok((bucket, None));
+		}
+	}
+
 	let (bucket, key) = match path.find('/') {
 		Some(i) => {
 			let key = &path[i + 1..];
@@ -278,7 +396,7 @@ mod tests {
 
 	#[test]
 	fn parse_bucket_containing_a_key() -> Result<(), Error> {
-		let (bucket, key) = parse_bucket_key("/my_bucket/a/super/file.jpg")?;
+		let (bucket, key) = parse_bucket_key("/my_bucket/a/super/file.jpg", None)?;
 		assert_eq!(bucket, "my_bucket");
 		assert_eq!(key.expect("key must be set"), "a/super/file.jpg");
 		Ok(())
@@ -286,10 +404,10 @@ mod tests {
 
 	#[test]
 	fn parse_bucket_containing_no_key() -> Result<(), Error> {
-		let (bucket, key) = parse_bucket_key("/my_bucket/")?;
+		let (bucket, key) = parse_bucket_key("/my_bucket/", None)?;
 		assert_eq!(bucket, "my_bucket");
 		assert!(key.is_none());
-		let (bucket, key) = parse_bucket_key("/my_bucket")?;
+		let (bucket, key) = parse_bucket_key("/my_bucket", None)?;
 		assert_eq!(bucket, "my_bucket");
 		assert!(key.is_none());
 		Ok(())
@@ -297,11 +415,30 @@ mod tests {
 
 	#[test]
 	fn parse_bucket_containing_no_bucket() {
-		let parsed = parse_bucket_key("");
+		let parsed = parse_bucket_key("", None);
 		assert!(parsed.is_err());
-		let parsed = parse_bucket_key("/");
+		let parsed = parse_bucket_key("/", None);
 		assert!(parsed.is_err());
-		let parsed = parse_bucket_key("////");
+		let parsed = parse_bucket_key("////", None);
 		assert!(parsed.is_err());
 	}
+
+	#[test]
+	fn parse_bucket_with_vhost_and_key() -> Result<(), Error> {
+		let (bucket, key) = parse_bucket_key("/a/super/file.jpg", Some("my-bucket"))?;
+		assert_eq!(bucket, "my-bucket");
+		assert_eq!(key.expect("key must be set"), "a/super/file.jpg");
+		Ok(())
+	}
+
+	#[test]
+	fn parse_bucket_with_vhost_no_key() -> Result<(), Error> {
+		let (bucket, key) = parse_bucket_key("", Some("my-bucket"))?;
+		assert_eq!(bucket, "my-bucket");
+		assert!(key.is_none());
+		let (bucket, key) = parse_bucket_key("/", Some("my-bucket"))?;
+		assert_eq!(bucket, "my-bucket");
+		assert!(key.is_none());
+		Ok(())
+	}
 }

src/api/error.rs

@@ -1,6 +1,10 @@
-use err_derive::Error;
-use hyper::StatusCode;
+use std::convert::TryInto;
 
+use err_derive::Error;
+use hyper::header::HeaderValue;
+use hyper::{HeaderMap, StatusCode};
+
+use garage_model::helper::error::Error as HelperError;
 use garage_util::error::Error as GarageError;
 
 use crate::s3_xml;
@@ -31,8 +35,28 @@ pub enum Error {
 	AuthorizationHeaderMalformed(String),
 
 	/// The object requested don't exists
-	#[error(display = "Not found")]
-	NotFound,
+	#[error(display = "Key not found")]
+	NoSuchKey,
+
+	/// The bucket requested don't exists
+	#[error(display = "Bucket not found")]
+	NoSuchBucket,
+
+	/// The multipart upload requested don't exists
+	#[error(display = "Upload not found")]
+	NoSuchUpload,
+
+	/// Tried to create a bucket that already exist
+	#[error(display = "Bucket already exists")]
+	BucketAlreadyExists,
+
+	/// Tried to delete a non-empty bucket
+	#[error(display = "Tried to delete a non-empty bucket")]
+	BucketNotEmpty,
+
+	/// Precondition failed (e.g. x-amz-copy-source-if-match)
+	#[error(display = "At least one of the preconditions you specified did not hold")]
+	PreconditionFailed,
 
 	// Category: bad request
 	/// The request contained an invalid UTF-8 sequence in its path or in other parameters
@@ -57,11 +81,15 @@ pub enum Error {
 
 	/// The client sent a range header with invalid value
 	#[error(display = "Invalid HTTP range: {:?}", _0)]
-	InvalidRange(#[error(from)] http_range::HttpRangeParseError),
+	InvalidRange(#[error(from)] (http_range::HttpRangeParseError, u64)),
 
 	/// The client sent an invalid request
 	#[error(display = "Bad request: {}", _0)]
 	BadRequest(String),
+
+	/// The client sent a request for an action not supported by garage
+	#[error(display = "Unimplemented action: {}", _0)]
+	NotImplemented(String),
 }
 
 impl From<roxmltree::Error> for Error {
@@ -76,26 +104,53 @@ impl From<quick_xml::de::DeError> for Error {
 	}
 }
 
+impl From<HelperError> for Error {
+	fn from(err: HelperError) -> Self {
+		match err {
+			HelperError::Internal(i) => Self::InternalError(i),
+			HelperError::BadRequest(b) => Self::BadRequest(b),
+		}
+	}
+}
+
 impl Error {
 	/// Get the HTTP status code that best represents the meaning of the error for the client
 	pub fn http_status_code(&self) -> StatusCode {
 		match self {
-			Error::NotFound => StatusCode::NOT_FOUND,
+			Error::NoSuchKey | Error::NoSuchBucket | Error::NoSuchUpload => StatusCode::NOT_FOUND,
+			Error::BucketNotEmpty | Error::BucketAlreadyExists => StatusCode::CONFLICT,
+			Error::PreconditionFailed => StatusCode::PRECONDITION_FAILED,
 			Error::Forbidden(_) => StatusCode::FORBIDDEN,
-			Error::InternalError(GarageError::Rpc(_)) => StatusCode::SERVICE_UNAVAILABLE,
+			Error::InternalError(
+				GarageError::Timeout
+				| GarageError::RemoteError(_)
+				| GarageError::Quorum(_, _, _, _),
+			) => StatusCode::SERVICE_UNAVAILABLE,
 			Error::InternalError(_) | Error::Hyper(_) | Error::Http(_) => {
 				StatusCode::INTERNAL_SERVER_ERROR
 			}
+			Error::InvalidRange(_) => StatusCode::RANGE_NOT_SATISFIABLE,
+			Error::NotImplemented(_) => StatusCode::NOT_IMPLEMENTED,
 			_ => StatusCode::BAD_REQUEST,
 		}
 	}
 
 	pub fn aws_code(&self) -> &'static str {
 		match self {
-			Error::NotFound => "NoSuchKey",
+			Error::NoSuchKey => "NoSuchKey",
+			Error::NoSuchBucket => "NoSuchBucket",
+			Error::NoSuchUpload => "NoSuchUpload",
+			Error::BucketAlreadyExists => "BucketAlreadyExists",
+			Error::BucketNotEmpty => "BucketNotEmpty",
+			Error::PreconditionFailed => "PreconditionFailed",
 			Error::Forbidden(_) => "AccessDenied",
 			Error::AuthorizationHeaderMalformed(_) => "AuthorizationHeaderMalformed",
-			Error::InternalError(GarageError::Rpc(_)) => "ServiceUnavailable",
+			Error::NotImplemented(_) => "NotImplemented",
+			Error::InternalError(
+				GarageError::Timeout
+				| GarageError::RemoteError(_)
+				| GarageError::Quorum(_, _, _, _),
+			) => "ServiceUnavailable",
 			Error::InternalError(_) | Error::Hyper(_) | Error::Http(_) => "InternalError",
 			_ => "InvalidRequest",
 		}
@@ -119,66 +174,87 @@ impl Error {
 			.into()
 		})
 	}
+
+	pub fn add_headers(&self, header_map: &mut HeaderMap<HeaderValue>) {
+		use hyper::header;
+		#[allow(clippy::single_match)]
+		match self {
+			Error::InvalidRange((_, len)) => {
+				header_map.append(
+					header::CONTENT_RANGE,
+					format!("bytes */{}", len)
+						.try_into()
+						.expect("header value only contain ascii"),
+				);
+			}
+			_ => (),
+		}
+	}
 }
 
 /// Trait to map error to the Bad Request error code
 pub trait OkOrBadRequest {
-	type S2;
-	fn ok_or_bad_request(self, reason: &'static str) -> Self::S2;
+	type S;
+	fn ok_or_bad_request<M: AsRef<str>>(self, reason: M) -> Result<Self::S, Error>;
 }
 
 impl<T, E> OkOrBadRequest for Result<T, E>
 where
 	E: std::fmt::Display,
 {
-	type S2 = Result<T, Error>;
-	fn ok_or_bad_request(self, reason: &'static str) -> Result<T, Error> {
+	type S = T;
+	fn ok_or_bad_request<M: AsRef<str>>(self, reason: M) -> Result<T, Error> {
 		match self {
 			Ok(x) => Ok(x),
-			Err(e) => Err(Error::BadRequest(format!("{}: {}", reason, e))),
+			Err(e) => Err(Error::BadRequest(format!(
+				"{}: {}",
+				reason.as_ref(),
+				e.to_string()
+			))),
 		}
 	}
 }
 
 impl<T> OkOrBadRequest for Option<T> {
-	type S2 = Result<T, Error>;
-	fn ok_or_bad_request(self, reason: &'static str) -> Result<T, Error> {
+	type S = T;
+	fn ok_or_bad_request<M: AsRef<str>>(self, reason: M) -> Result<T, Error> {
 		match self {
 			Some(x) => Ok(x),
-			None => Err(Error::BadRequest(reason.to_string())),
+			None => Err(Error::BadRequest(reason.as_ref().to_string())),
 		}
 	}
 }
 
 /// Trait to map an error to an Internal Error code
 pub trait OkOrInternalError {
-	type S2;
-	fn ok_or_internal_error(self, reason: &'static str) -> Self::S2;
+	type S;
+	fn ok_or_internal_error<M: AsRef<str>>(self, reason: M) -> Result<Self::S, Error>;
 }
 
 impl<T, E> OkOrInternalError for Result<T, E>
 where
 	E: std::fmt::Display,
 {
-	type S2 = Result<T, Error>;
-	fn ok_or_internal_error(self, reason: &'static str) -> Result<T, Error> {
+	type S = T;
+	fn ok_or_internal_error<M: AsRef<str>>(self, reason: M) -> Result<T, Error> {
 		match self {
 			Ok(x) => Ok(x),
 			Err(e) => Err(Error::InternalError(GarageError::Message(format!(
 				"{}: {}",
-				reason, e
+				reason.as_ref(),
+				e
 			)))),
 		}
 	}
 }
 
 impl<T> OkOrInternalError for Option<T> {
-	type S2 = Result<T, Error>;
-	fn ok_or_internal_error(self, reason: &'static str) -> Result<T, Error> {
+	type S = T;
+	fn ok_or_internal_error<M: AsRef<str>>(self, reason: M) -> Result<T, Error> {
 		match self {
 			Some(x) => Ok(x),
 			None => Err(Error::InternalError(GarageError::Message(
-				reason.to_string(),
+				reason.as_ref().to_string(),
 			))),
 		}
 	}

src/api/helpers.rs

@@ -0,0 +1,114 @@
+use crate::Error;
+use idna::domain_to_unicode;
+
+/// Host to bucket
+///
+/// Convert a host, like "bucket.garage-site.tld" to the corresponding bucket "bucket",
+/// considering that ".garage-site.tld" is the "root domain". For domains not matching
+/// the provided root domain, no bucket is returned
+/// This behavior has been chosen to follow AWS S3 semantic.
+pub fn host_to_bucket<'a>(host: &'a str, root: &str) -> Option<&'a str> {
+	let root = root.trim_start_matches('.');
+	let label_root = root.chars().filter(|c| c == &'.').count() + 1;
+	let root = root.rsplit('.');
+	let mut host = host.rsplitn(label_root + 1, '.');
+	for root_part in root {
+		let host_part = host.next()?;
+		if root_part != host_part {
+			return None;
+		}
+	}
+	host.next()
+}
+
+/// Extract host from the authority section given by the HTTP host header
+///
+/// The HTTP host contains both a host and a port.
+/// Extracting the port is more complex than just finding the colon (:) symbol due to IPv6
+/// We do not use the collect pattern as there is no way in std rust to collect over a stack allocated value
+/// check here: <https://docs.rs/collect_slice/1.2.0/collect_slice/>
+pub fn authority_to_host(authority: &str) -> Result<String, Error> {
+	let mut iter = authority.chars().enumerate();
+	let (_, first_char) = iter
+		.next()
+		.ok_or_else(|| Error::BadRequest("Authority is empty".to_string()))?;
+
+	let split = match first_char {
+		'[' => {
+			let mut iter = iter.skip_while(|(_, c)| c != &']');
+			match iter.next() {
+				Some((_, ']')) => iter.next(),
+				_ => {
+					return Err(Error::BadRequest(format!(
+						"Authority {} has an illegal format",
+						authority
+					)))
+				}
+			}
+		}
+		_ => iter.find(|(_, c)| *c == ':'),
+	};
+
+	let authority = match split {
+		Some((i, ':')) => Ok(&authority[..i]),
+		None => Ok(authority),
+		Some((_, _)) => Err(Error::BadRequest(format!(
+			"Authority {} has an illegal format",
+			authority
+		))),
+	};
+	authority.map(|h| domain_to_unicode(h).0)
+}
+
+#[cfg(test)]
+mod tests {
+	use super::*;
+
+	#[test]
+	fn authority_to_host_with_port() -> Result<(), Error> {
+		let domain = authority_to_host("[::1]:3902")?;
+		assert_eq!(domain, "[::1]");
+		let domain2 = authority_to_host("garage.tld:65200")?;
+		assert_eq!(domain2, "garage.tld");
+		let domain3 = authority_to_host("127.0.0.1:80")?;
+		assert_eq!(domain3, "127.0.0.1");
+		Ok(())
+	}
+
+	#[test]
+	fn authority_to_host_without_port() -> Result<(), Error> {
+		let domain = authority_to_host("[::1]")?;
+		assert_eq!(domain, "[::1]");
+		let domain2 = authority_to_host("garage.tld")?;
+		assert_eq!(domain2, "garage.tld");
+		let domain3 = authority_to_host("127.0.0.1")?;
+		assert_eq!(domain3, "127.0.0.1");
+		assert!(authority_to_host("[").is_err());
+		assert!(authority_to_host("[hello").is_err());
+		Ok(())
+	}
+
+	#[test]
+	fn host_to_bucket_test() {
+		assert_eq!(
+			host_to_bucket("john.doe.garage.tld", ".garage.tld").unwrap(),
+			"john.doe"
+		);
+
+		assert_eq!(
+			host_to_bucket("john.doe.garage.tld", "garage.tld").unwrap(),
+			"john.doe"
+		);
+
+		assert_eq!(host_to_bucket("john.doe.com", "garage.tld"), None);
+
+		assert_eq!(host_to_bucket("john.doe.com", ".garage.tld"), None);
+
+		assert_eq!(host_to_bucket("garage.tld", "garage.tld"), None);
+
+		assert_eq!(host_to_bucket("garage.tld", ".garage.tld"), None);
+
+		assert_eq!(host_to_bucket("not-garage.tld", "garage.tld"), None);
+		assert_eq!(host_to_bucket("not-garage.tld", ".garage.tld"), None);
+	}
+}

src/api/lib.rs

@@ -2,7 +2,7 @@
 #[macro_use]
 extern crate log;
 
-mod error;
+pub mod error;
 pub use error::Error;
 
 mod encoding;
@@ -12,10 +12,14 @@ pub use api_server::run_api_server;
 
 mod signature;
 
+pub mod helpers;
 mod s3_bucket;
 mod s3_copy;
+pub mod s3_cors;
 mod s3_delete;
 pub mod s3_get;
 mod s3_list;
 mod s3_put;
+mod s3_router;
+mod s3_website;
 mod s3_xml;

src/api/s3_bucket.rs

@@ -1,13 +1,22 @@
+use std::collections::HashMap;
 use std::sync::Arc;
 
-use hyper::{Body, Response};
+use hyper::{Body, Request, Response, StatusCode};
 
+use garage_model::bucket_alias_table::*;
+use garage_model::bucket_table::Bucket;
 use garage_model::garage::Garage;
 use garage_model::key_table::Key;
+use garage_model::object_table::ObjectFilter;
+use garage_model::permission::BucketKeyPerm;
+use garage_table::util::*;
+use garage_util::crdt::*;
+use garage_util::data::*;
 use garage_util::time::*;
 
 use crate::error::*;
 use crate::s3_xml;
+use crate::signature::verify_signed_content;
 
 pub fn handle_get_bucket_location(garage: Arc<Garage>) -> Result<Response<Body>, Error> {
 	let loc = s3_xml::LocationConstraint {
@@ -34,19 +43,61 @@ pub fn handle_get_bucket_versioning() -> Result<Response<Body>, Error> {
 		.body(Body::from(xml.into_bytes()))?)
 }
 
-pub fn handle_list_buckets(api_key: &Key) -> Result<Response<Body>, Error> {
+pub async fn handle_list_buckets(garage: &Garage, api_key: &Key) -> Result<Response<Body>, Error> {
+	let key_p = api_key.params().ok_or_internal_error(
+		"Key should not be in deleted state at this point (in handle_list_buckets)",
+	)?;
+
+	// Collect buckets user has access to
+	let ids = api_key
+		.state
+		.as_option()
+		.unwrap()
+		.authorized_buckets
+		.items()
+		.iter()
+		.filter(|(_, perms)| perms.is_any())
+		.map(|(id, _)| *id)
+		.collect::<Vec<_>>();
+
+	let mut buckets_by_id = HashMap::new();
+	let mut aliases = HashMap::new();
+
+	for bucket_id in ids.iter() {
+		let bucket = garage.bucket_table.get(&EmptyKey, bucket_id).await?;
+		if let Some(bucket) = bucket {
+			for (alias, _, _active) in bucket.aliases().iter().filter(|(_, _, active)| *active) {
+				let alias_opt = garage.bucket_alias_table.get(&EmptyKey, alias).await?;
+				if let Some(alias_ent) = alias_opt {
+					if *alias_ent.state.get() == Some(*bucket_id) {
+						aliases.insert(alias_ent.name().to_string(), *bucket_id);
+					}
+				}
+			}
+			if let Deletable::Present(param) = bucket.state {
+				buckets_by_id.insert(bucket_id, param);
+			}
+		}
+	}
+
+	for (alias, _, id_opt) in key_p.local_aliases.items() {
+		if let Some(id) = id_opt {
+			aliases.insert(alias.clone(), *id);
+		}
+	}
+
+	// Generate response
 	let list_buckets = s3_xml::ListAllMyBucketsResult {
 		owner: s3_xml::Owner {
-			display_name: s3_xml::Value(api_key.name.get().to_string()),
+			display_name: s3_xml::Value(key_p.name.get().to_string()),
 			id: s3_xml::Value(api_key.key_id.to_string()),
 		},
 		buckets: s3_xml::BucketList {
-			entries: api_key
-				.authorized_buckets
-				.items()
+			entries: aliases
 				.iter()
-				.map(|(name, ts, _)| s3_xml::Bucket {
-					creation_date: s3_xml::Value(msec_to_rfc3339(*ts)),
+				.filter_map(|(name, id)| buckets_by_id.get(id).map(|p| (name, id, p)))
+				.map(|(name, _id, param)| s3_xml::Bucket {
+					creation_date: s3_xml::Value(msec_to_rfc3339(param.creation_date)),
 					name: s3_xml::Value(name.to_string()),
 				})
 				.collect(),
@@ -60,3 +111,239 @@ pub fn handle_list_buckets(api_key: &Key) -> Result<Response<Body>, Error> {
 		.header("Content-Type", "application/xml")
 		.body(Body::from(xml))?)
 }
+
+pub async fn handle_create_bucket(
+	garage: &Garage,
+	req: Request<Body>,
+	content_sha256: Option<Hash>,
+	api_key: Key,
+	bucket_name: String,
+) -> Result<Response<Body>, Error> {
+	let body = hyper::body::to_bytes(req.into_body()).await?;
+	verify_signed_content(content_sha256, &body[..])?;
+
+	let cmd =
+		parse_create_bucket_xml(&body[..]).ok_or_bad_request("Invalid create bucket XML query")?;
+
+	if let Some(location_constraint) = cmd {
+		if location_constraint != garage.config.s3_api.s3_region {
+			return Err(Error::BadRequest(format!(
+				"Cannot satisfy location constraint `{}`: buckets can only be created in region `{}`",
+				location_constraint,
+				garage.config.s3_api.s3_region
+			)));
+		}
+	}
+
+	let key_params = api_key
+		.params()
+		.ok_or_internal_error("Key should not be deleted at this point")?;
+
+	let existing_bucket = if let Some(Some(bucket_id)) = key_params.local_aliases.get(&bucket_name)
+	{
+		Some(*bucket_id)
+	} else {
+		garage
+			.bucket_helper()
+			.resolve_global_bucket_name(&bucket_name)
+			.await?
+	};
+
+	if let Some(bucket_id) = existing_bucket {
+		// Check we have write or owner permission on the bucket,
+		// in that case it's fine, return 200 OK, bucket exists;
+		// otherwise return a forbidden error.
+		let kp = api_key.bucket_permissions(&bucket_id);
+		if !(kp.allow_write || kp.allow_owner) {
+			return Err(Error::BucketAlreadyExists);
+		}
+	} else {
+		// Create the bucket!
+		if !is_valid_bucket_name(&bucket_name) {
+			return Err(Error::BadRequest(format!(
+				"{}: {}",
+				bucket_name, INVALID_BUCKET_NAME_MESSAGE
+			)));
+		}
+
+		let bucket = Bucket::new();
+		garage.bucket_table.insert(&bucket).await?;
+
+		garage
+			.bucket_helper()
+			.set_bucket_key_permissions(bucket.id, &api_key.key_id, BucketKeyPerm::ALL_PERMISSIONS)
+			.await?;
+
+		garage
+			.bucket_helper()
+			.set_local_bucket_alias(bucket.id, &api_key.key_id, &bucket_name)
+			.await?;
+	}
+
+	Ok(Response::builder()
+		.header("Location", format!("/{}", bucket_name))
+		.body(Body::empty())
+		.unwrap())
+}
+
+pub async fn handle_delete_bucket(
+	garage: &Garage,
+	bucket_id: Uuid,
+	bucket_name: String,
+	api_key: Key,
+) -> Result<Response<Body>, Error> {
+	let key_params = api_key
+		.params()
+		.ok_or_internal_error("Key should not be deleted at this point")?;
+
+	let is_local_alias = matches!(key_params.local_aliases.get(&bucket_name), Some(Some(_)));
+
+	let mut bucket = garage
+		.bucket_helper()
+		.get_existing_bucket(bucket_id)
+		.await?;
+	let bucket_state = bucket.state.as_option().unwrap();
+
+	// If the bucket has no other aliases, this is a true deletion.
+	// Otherwise, it is just an alias removal.
+
+	let has_other_global_aliases = bucket_state
+		.aliases
+		.items()
+		.iter()
+		.filter(|(_, _, active)| *active)
+		.any(|(n, _, _)| is_local_alias || (*n != bucket_name));
+
+	let has_other_local_aliases = bucket_state
+		.local_aliases
+		.items()
+		.iter()
+		.filter(|(_, _, active)| *active)
+		.any(|((k, n), _, _)| !is_local_alias || *n != bucket_name || *k != api_key.key_id);
+
+	if !has_other_global_aliases && !has_other_local_aliases {
+		// Delete bucket
+
+		// Check bucket is empty
+		let objects = garage
+			.object_table
+			.get_range(&bucket_id, None, Some(ObjectFilter::IsData), 10)
+			.await?;
+		if !objects.is_empty() {
+			return Err(Error::BucketNotEmpty);
+		}
+
+		// --- done checking, now commit ---
+		// 1. delete bucket alias
+		if is_local_alias {
+			garage
+				.bucket_helper()
+				.unset_local_bucket_alias(bucket_id, &api_key.key_id, &bucket_name)
+				.await?;
+		} else {
+			garage
+				.bucket_helper()
+				.unset_global_bucket_alias(bucket_id, &bucket_name)
+				.await?;
+		}
+
+		// 2. delete authorization from keys that had access
+		for (key_id, _) in bucket.authorized_keys() {
+			garage
+				.bucket_helper()
+				.set_bucket_key_permissions(bucket.id, key_id, BucketKeyPerm::NO_PERMISSIONS)
+				.await?;
+		}
+
+		// 3. delete bucket
+		bucket.state = Deletable::delete();
+		garage.bucket_table.insert(&bucket).await?;
+	} else if is_local_alias {
+		// Just unalias
+		garage
+			.bucket_helper()
+			.unset_local_bucket_alias(bucket_id, &api_key.key_id, &bucket_name)
+			.await?;
+	} else {
+		// Just unalias (but from global namespace)
+		garage
+			.bucket_helper()
+			.unset_global_bucket_alias(bucket_id, &bucket_name)
+			.await?;
+	}
+
+	Ok(Response::builder()
+		.status(StatusCode::NO_CONTENT)
+		.body(Body::empty())?)
+}
+
+fn parse_create_bucket_xml(xml_bytes: &[u8]) -> Option<Option<String>> {
+	// Returns None if invalid data
+	// Returns Some(None) if no location constraint is given
+	// Returns Some(Some("xxxx")) where xxxx is the given location constraint
+
+	let xml_str = std::str::from_utf8(xml_bytes).ok()?;
+	if xml_str.trim_matches(char::is_whitespace).is_empty() {
+		return Some(None);
+	}
+
+	let xml = roxmltree::Document::parse(xml_str).ok()?;
+
+	let cbc = xml.root().first_child()?;
+	if !cbc.has_tag_name("CreateBucketConfiguration") {
+		return None;
+	}
+
+	let mut ret = None;
+	for item in cbc.children() {
+		println!("{:?}", item);
+		if item.has_tag_name("LocationConstraint") {
+			if ret != None {
+				return None;
+			}
+			ret = Some(item.text()?.to_string());
+		} else if !item.is_text() {
+			return None;
+		}
+	}
+
+	Some(ret)
+}
+
+#[cfg(test)]
+mod tests {
+	use super::*;
+
+	#[test]
+	fn create_bucket() {
+		assert_eq!(parse_create_bucket_xml(br#""#), Some(None));
+		assert_eq!(
+			parse_create_bucket_xml(
+				br#"
+            <CreateBucketConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/"> 
+            </CreateBucketConfiguration >
+		"#
+			),
+			Some(None)
+		);
+		assert_eq!(
+			parse_create_bucket_xml(
+				br#"
+            <CreateBucketConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/"> 
+             <LocationConstraint>Europe</LocationConstraint> 
+            </CreateBucketConfiguration >
+		"#
+			),
+			Some(Some("Europe".into()))
+		);
+		assert_eq!(
+			parse_create_bucket_xml(
+				br#"
+            <CreateBucketConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/"> 
+            </Crea >
+		"#
+			),
+			None
+		);
+	}
+}

src/api/s3_copy.rs

@@ -1,6 +1,11 @@
 use std::sync::Arc;
+use std::time::{Duration, SystemTime, UNIX_EPOCH};
+
+use futures::TryFutureExt;
+use md5::{Digest as Md5Digest, Md5};
 
 use hyper::{Body, Request, Response};
+use serde::Serialize;
 
 use garage_table::*;
 use garage_util::data::*;
@@ -8,63 +13,50 @@ use garage_util::time::*;
 
 use garage_model::block_ref_table::*;
 use garage_model::garage::Garage;
+use garage_model::key_table::Key;
 use garage_model::object_table::*;
 use garage_model::version_table::*;
 
+use crate::api_server::{parse_bucket_key, resolve_bucket};
 use crate::error::*;
-use crate::s3_put::get_headers;
-use crate::s3_xml;
+use crate::s3_put::{decode_upload_id, get_headers};
+use crate::s3_xml::{self, xmlns_tag};
 
 pub async fn handle_copy(
 	garage: Arc<Garage>,
+	api_key: &Key,
 	req: &Request<Body>,
-	dest_bucket: &str,
+	dest_bucket_id: Uuid,
 	dest_key: &str,
-	source_bucket: &str,
-	source_key: &str,
 ) -> Result<Response<Body>, Error> {
-	let source_object = garage
-		.object_table
-		.get(&source_bucket.to_string(), &source_key.to_string())
-		.await?
-		.ok_or(Error::NotFound)?;
+	let copy_precondition = CopyPreconditionHeaders::parse(req)?;
 
-	let source_last_v = source_object
-		.versions()
-		.iter()
-		.rev()
-		.find(|v| v.is_complete())
-		.ok_or(Error::NotFound)?;
+	let source_object = get_copy_source(&garage, api_key, req).await?;
 
-	let source_last_state = match &source_last_v.state {
-		ObjectVersionState::Complete(x) => x,
-		_ => unreachable!(),
-	};
+	let (source_version, source_version_data, source_version_meta) =
+		extract_source_info(&source_object)?;
 
+	// Check precondition, e.g. x-amz-copy-source-if-match
+	copy_precondition.check(source_version, &source_version_meta.etag)?;
+
+	// Generate parameters for copied object
 	let new_uuid = gen_uuid();
 	let new_timestamp = now_msec();
 
 	// Implement x-amz-metadata-directive: REPLACE
-	let old_meta = match source_last_state {
-		ObjectVersionData::DeleteMarker => {
-			return Err(Error::NotFound);
-		}
-		ObjectVersionData::Inline(meta, _bytes) => meta,
-		ObjectVersionData::FirstBlock(meta, _fbh) => meta,
-	};
 	let new_meta = match req.headers().get("x-amz-metadata-directive") {
 		Some(v) if v == hyper::header::HeaderValue::from_static("REPLACE") => ObjectVersionMeta {
 			headers: get_headers(req)?,
-			size: old_meta.size,
-			etag: old_meta.etag.clone(),
+			size: source_version_meta.size,
+			etag: source_version_meta.etag.clone(),
 		},
-		_ => old_meta.clone(),
+		_ => source_version_meta.clone(),
 	};
 
 	let etag = new_meta.etag.to_string();
 
 	// Save object copy
-	match source_last_state {
+	match source_version_data {
 		ObjectVersionData::DeleteMarker => unreachable!(),
 		ObjectVersionData::Inline(_meta, bytes) => {
 			let dest_object_version = ObjectVersion {
@@ -76,7 +68,7 @@ pub async fn handle_copy(
 				)),
 			};
 			let dest_object = Object::new(
-				dest_bucket.to_string(),
+				dest_bucket_id,
 				dest_key.to_string(),
 				vec![dest_object_version],
 			);
@@ -86,9 +78,9 @@ pub async fn handle_copy(
 			// Get block list from source version
 			let source_version = garage
 				.version_table
-				.get(&source_last_v.uuid, &EmptyKey)
+				.get(&source_version.uuid, &EmptyKey)
 				.await?;
-			let source_version = source_version.ok_or(Error::NotFound)?;
+			let source_version = source_version.ok_or(Error::NoSuchKey)?;
 
 			// Write an "uploading" marker in Object table
 			// This holds a reference to the object in the Version table
@@ -99,7 +91,7 @@ pub async fn handle_copy(
 				state: ObjectVersionState::Uploading(new_meta.headers.clone()),
 			};
 			let tmp_dest_object = Object::new(
-				dest_bucket.to_string(),
+				dest_bucket_id,
 				dest_key.to_string(),
 				vec![tmp_dest_object_version],
 			);
@@ -109,12 +101,8 @@ pub async fn handle_copy(
 			// this means that the BlockRef entries linked to this version cannot be
 			// marked as deleted (they are marked as deleted only if the Version
 			// doesn't exist or is marked as deleted).
-			let mut dest_version = Version::new(
-				new_uuid,
-				dest_bucket.to_string(),
-				dest_key.to_string(),
-				false,
-			);
+			let mut dest_version =
+				Version::new(new_uuid, dest_bucket_id, dest_key.to_string(), false);
 			garage.version_table.insert(&dest_version).await?;
 
 			// Fill in block list for version and insert block refs
@@ -151,7 +139,7 @@ pub async fn handle_copy(
 				)),
 			};
 			let dest_object = Object::new(
-				dest_bucket.to_string(),
+				dest_bucket_id,
 				dest_key.to_string(),
 				vec![dest_object_version],
 			);
@@ -160,13 +148,468 @@ pub async fn handle_copy(
 	}
 
 	let last_modified = msec_to_rfc3339(new_timestamp);
-	let result = s3_xml::CopyObjectResult {
+	let result = CopyObjectResult {
 		last_modified: s3_xml::Value(last_modified),
-		etag: s3_xml::Value(etag),
+		etag: s3_xml::Value(format!("\"{}\"", etag)),
 	};
 	let xml = s3_xml::to_xml_with_header(&result)?;
 
 	Ok(Response::builder()
 		.header("Content-Type", "application/xml")
+		.header("x-amz-version-id", hex::encode(new_uuid))
+		.header(
+			"x-amz-copy-source-version-id",
+			hex::encode(source_version.uuid),
+		)
 		.body(Body::from(xml))?)
 }
+
+pub async fn handle_upload_part_copy(
+	garage: Arc<Garage>,
+	api_key: &Key,
+	req: &Request<Body>,
+	dest_bucket_id: Uuid,
+	dest_key: &str,
+	part_number: u64,
+	upload_id: &str,
+) -> Result<Response<Body>, Error> {
+	let copy_precondition = CopyPreconditionHeaders::parse(req)?;
+
+	let dest_version_uuid = decode_upload_id(upload_id)?;
+
+	let dest_key = dest_key.to_string();
+	let (source_object, dest_object) = futures::try_join!(
+		get_copy_source(&garage, api_key, req),
+		garage
+			.object_table
+			.get(&dest_bucket_id, &dest_key)
+			.map_err(Error::from),
+	)?;
+	let dest_object = dest_object.ok_or(Error::NoSuchKey)?;
+
+	let (source_object_version, source_version_data, source_version_meta) =
+		extract_source_info(&source_object)?;
+
+	// Check precondition on source, e.g. x-amz-copy-source-if-match
+	copy_precondition.check(source_object_version, &source_version_meta.etag)?;
+
+	// Check source range is valid
+	let source_range = match req.headers().get("x-amz-copy-source-range") {
+		Some(range) => {
+			let range_str = range.to_str()?;
+			let mut ranges = http_range::HttpRange::parse(range_str, source_version_meta.size)
+				.map_err(|e| (e, source_version_meta.size))?;
+			if ranges.len() != 1 {
+				return Err(Error::BadRequest(
+					"Invalid x-amz-copy-source-range header: exactly 1 range must be given".into(),
+				));
+			} else {
+				ranges.pop().unwrap()
+			}
+		}
+		None => http_range::HttpRange {
+			start: 0,
+			length: source_version_meta.size,
+		},
+	};
+
+	// Check destination version is indeed in uploading state
+	if !dest_object
+		.versions()
+		.iter()
+		.any(|v| v.uuid == dest_version_uuid && v.is_uploading())
+	{
+		return Err(Error::NoSuchUpload);
+	}
+
+	// Check source version is not inlined
+	match source_version_data {
+		ObjectVersionData::DeleteMarker => unreachable!(),
+		ObjectVersionData::Inline(_meta, _bytes) => {
+			// This is only for small files, we don't bother handling this.
+			// (in AWS UploadPartCopy works for parts at least 5MB which
+			// is never the case of an inline object)
+			return Err(Error::BadRequest(
+				"Source object is too small (minimum part size is 5Mb)".into(),
+			));
+		}
+		ObjectVersionData::FirstBlock(_meta, _first_block_hash) => (),
+	};
+
+	// Fetch source versin with its block list,
+	// and destination version to check part hasn't yet been uploaded
+	let (source_version, dest_version) = futures::try_join!(
+		garage
+			.version_table
+			.get(&source_object_version.uuid, &EmptyKey),
+		garage.version_table.get(&dest_version_uuid, &EmptyKey),
+	)?;
+	let source_version = source_version.ok_or(Error::NoSuchKey)?;
+
+	// Check this part number hasn't yet been uploaded
+	if let Some(dv) = dest_version {
+		if dv.has_part_number(part_number) {
+			return Err(Error::BadRequest(format!(
+				"Part number {} has already been uploaded",
+				part_number
+			)));
+		}
+	}
+
+	// We want to reuse blocks from the source version as much as possible.
+	// However, we still need to get the data from these blocks
+	// because we need to know it to calculate the MD5sum of the part
+	// which is used as its ETag.
+
+	// First, calculate what blocks we want to keep,
+	// and the subrange of the block to take, if the bounds of the
+	// requested range are in the middle.
+	let (range_begin, range_end) = (source_range.start, source_range.start + source_range.length);
+
+	let mut blocks_to_copy = vec![];
+	let mut current_offset = 0;
+	let mut size_to_copy = 0;
+	for (_bk, block) in source_version.blocks.items().iter() {
+		let (block_begin, block_end) = (current_offset, current_offset + block.size);
+
+		if block_begin < range_end && block_end > range_begin {
+			let subrange_begin = if block_begin < range_begin {
+				Some(range_begin - block_begin)
+			} else {
+				None
+			};
+			let subrange_end = if block_end > range_end {
+				Some(range_end - block_begin)
+			} else {
+				None
+			};
+			let range_to_copy = match (subrange_begin, subrange_end) {
+				(Some(b), Some(e)) => Some(b as usize..e as usize),
+				(None, Some(e)) => Some(0..e as usize),
+				(Some(b), None) => Some(b as usize..block.size as usize),
+				(None, None) => None,
+			};
+			size_to_copy += range_to_copy
+				.as_ref()
+				.map(|x| x.len() as u64)
+				.unwrap_or(block.size);
+
+			blocks_to_copy.push((block.hash, range_to_copy));
+		}
+
+		current_offset = block_end;
+	}
+
+	if size_to_copy < 1024 * 1024 {
+		return Err(Error::BadRequest(format!(
+			"Not enough data to copy: {} bytes (minimum: 1MB)",
+			size_to_copy
+		)));
+	}
+
+	// Now, actually copy the blocks
+	let mut md5hasher = Md5::new();
+
+	let mut block = Some(
+		garage
+			.block_manager
+			.rpc_get_block(&blocks_to_copy[0].0)
+			.await?,
+	);
+
+	let mut current_offset = 0;
+	for (i, (block_hash, range_to_copy)) in blocks_to_copy.iter().enumerate() {
+		let (current_block, subrange_hash) = match range_to_copy.clone() {
+			Some(r) => {
+				let subrange = block.take().unwrap()[r].to_vec();
+				let hash = blake2sum(&subrange);
+				(subrange, hash)
+			}
+			None => (block.take().unwrap(), *block_hash),
+		};
+		md5hasher.update(&current_block[..]);
+
+		let mut version = Version::new(dest_version_uuid, dest_bucket_id, dest_key.clone(), false);
+		version.blocks.put(
+			VersionBlockKey {
+				part_number,
+				offset: current_offset,
+			},
+			VersionBlock {
+				hash: subrange_hash,
+				size: current_block.len() as u64,
+			},
+		);
+		current_offset += current_block.len() as u64;
+
+		let block_ref = BlockRef {
+			block: subrange_hash,
+			version: dest_version_uuid,
+			deleted: false.into(),
+		};
+
+		let next_block_hash = blocks_to_copy.get(i + 1).map(|(h, _)| *h);
+
+		let garage2 = garage.clone();
+		let garage3 = garage.clone();
+		let is_subrange = range_to_copy.is_some();
+
+		let (_, _, _, next_block) = futures::try_join!(
+			// Thing 1: if we are taking a subrange of the source block,
+			// we need to insert that subrange as a new block.
+			async move {
+				if is_subrange {
+					garage2
+						.block_manager
+						.rpc_put_block(subrange_hash, current_block)
+						.await
+				} else {
+					Ok(())
+				}
+			},
+			// Thing 2: we need to insert the block in the version
+			garage.version_table.insert(&version),
+			// Thing 3: we need to add a block reference
+			garage.block_ref_table.insert(&block_ref),
+			// Thing 4: we need to prefetch the next block
+			async move {
+				match next_block_hash {
+					Some(h) => Ok(Some(garage3.block_manager.rpc_get_block(&h).await?)),
+					None => Ok(None),
+				}
+			},
+		)?;
+
+		block = next_block;
+	}
+
+	let data_md5sum = md5hasher.finalize();
+	let etag = hex::encode(data_md5sum);
+
+	// Put the part's ETag in the Versiontable
+	let mut version = Version::new(dest_version_uuid, dest_bucket_id, dest_key.clone(), false);
+	version.parts_etags.put(part_number, etag.clone());
+	garage.version_table.insert(&version).await?;
+
+	// LGTM
+	let resp_xml = s3_xml::to_xml_with_header(&CopyPartResult {
+		xmlns: (),
+		etag: s3_xml::Value(format!("\"{}\"", etag)),
+		last_modified: s3_xml::Value(msec_to_rfc3339(source_object_version.timestamp)),
+	})?;
+
+	Ok(Response::builder()
+		.header("Content-Type", "application/xml")
+		.header(
+			"x-amz-copy-source-version-id",
+			hex::encode(source_object_version.uuid),
+		)
+		.body(Body::from(resp_xml))?)
+}
+
+async fn get_copy_source(
+	garage: &Garage,
+	api_key: &Key,
+	req: &Request<Body>,
+) -> Result<Object, Error> {
+	let copy_source = req.headers().get("x-amz-copy-source").unwrap().to_str()?;
+	let copy_source = percent_encoding::percent_decode_str(copy_source).decode_utf8()?;
+
+	let (source_bucket, source_key) = parse_bucket_key(&copy_source, None)?;
+	let source_bucket_id = resolve_bucket(garage, &source_bucket.to_string(), api_key).await?;
+
+	if !api_key.allow_read(&source_bucket_id) {
+		return Err(Error::Forbidden(format!(
+			"Reading from bucket {} not allowed for this key",
+			source_bucket
+		)));
+	}
+
+	let source_key = source_key.ok_or_bad_request("No source key specified")?;
+
+	let source_object = garage
+		.object_table
+		.get(&source_bucket_id, &source_key.to_string())
+		.await?
+		.ok_or(Error::NoSuchKey)?;
+
+	Ok(source_object)
+}
+
+fn extract_source_info(
+	source_object: &Object,
+) -> Result<(&ObjectVersion, &ObjectVersionData, &ObjectVersionMeta), Error> {
+	let source_version = source_object
+		.versions()
+		.iter()
+		.rev()
+		.find(|v| v.is_complete())
+		.ok_or(Error::NoSuchKey)?;
+
+	let source_version_data = match &source_version.state {
+		ObjectVersionState::Complete(x) => x,
+		_ => unreachable!(),
+	};
+
+	let source_version_meta = match source_version_data {
+		ObjectVersionData::DeleteMarker => {
+			return Err(Error::NoSuchKey);
+		}
+		ObjectVersionData::Inline(meta, _bytes) => meta,
+		ObjectVersionData::FirstBlock(meta, _fbh) => meta,
+	};
+
+	Ok((source_version, source_version_data, source_version_meta))
+}
+
+struct CopyPreconditionHeaders {
+	copy_source_if_match: Option<Vec<String>>,
+	copy_source_if_modified_since: Option<SystemTime>,
+	copy_source_if_none_match: Option<Vec<String>>,
+	copy_source_if_unmodified_since: Option<SystemTime>,
+}
+
+impl CopyPreconditionHeaders {
+	fn parse(req: &Request<Body>) -> Result<Self, Error> {
+		Ok(Self {
+			copy_source_if_match: req
+				.headers()
+				.get("x-amz-copy-source-if-match")
+				.map(|x| x.to_str())
+				.transpose()?
+				.map(|x| {
+					x.split(',')
+						.map(|m| m.trim().trim_matches('"').to_string())
+						.collect::<Vec<_>>()
+				}),
+			copy_source_if_modified_since: req
+				.headers()
+				.get("x-amz-copy-source-if-modified-since")
+				.map(|x| x.to_str())
+				.transpose()?
+				.map(|x| httpdate::parse_http_date(x))
+				.transpose()
+				.ok_or_bad_request("Invalid date in x-amz-copy-source-if-modified-since")?,
+			copy_source_if_none_match: req
+				.headers()
+				.get("x-amz-copy-source-if-none-match")
+				.map(|x| x.to_str())
+				.transpose()?
+				.map(|x| {
+					x.split(',')
+						.map(|m| m.trim().trim_matches('"').to_string())
+						.collect::<Vec<_>>()
+				}),
+			copy_source_if_unmodified_since: req
+				.headers()
+				.get("x-amz-copy-source-if-unmodified-since")
+				.map(|x| x.to_str())
+				.transpose()?
+				.map(|x| httpdate::parse_http_date(x))
+				.transpose()
+				.ok_or_bad_request("Invalid date in x-amz-copy-source-if-unmodified-since")?,
+		})
+	}
+
+	fn check(&self, v: &ObjectVersion, etag: &str) -> Result<(), Error> {
+		let v_date = UNIX_EPOCH + Duration::from_millis(v.timestamp);
+
+		let ok = match (
+			&self.copy_source_if_match,
+			&self.copy_source_if_unmodified_since,
+			&self.copy_source_if_none_match,
+			&self.copy_source_if_modified_since,
+		) {
+			// TODO I'm not sure all of the conditions are evaluated correctly here
+
+			// If we have both if-match and if-unmodified-since,
+			// basically we don't care about if-unmodified-since,
+			// because in the spec it says that if if-match evaluates to
+			// true but if-unmodified-since evaluates to false,
+			// the copy is still done.
+			(Some(im), _, None, None) => im.iter().any(|x| x == etag || x == "*"),
+			(None, Some(ius), None, None) => v_date <= *ius,
+
+			// If we have both if-none-match and if-modified-since,
+			// then both of the two conditions must evaluate to true
+			(None, None, Some(inm), Some(ims)) => {
+				!inm.iter().any(|x| x == etag || x == "*") && v_date > *ims
+			}
+			(None, None, Some(inm), None) => !inm.iter().any(|x| x == etag || x == "*"),
+			(None, None, None, Some(ims)) => v_date > *ims,
+			(None, None, None, None) => true,
+			_ => {
+				return Err(Error::BadRequest(
+					"Invalid combination of x-amz-copy-source-if-xxxxx headers".into(),
+				))
+			}
+		};
+
+		if ok {
+			Ok(())
+		} else {
+			Err(Error::PreconditionFailed)
+		}
+	}
+}
+
+#[derive(Debug, Serialize, PartialEq)]
+pub struct CopyObjectResult {
+	#[serde(rename = "LastModified")]
+	pub last_modified: s3_xml::Value,
+	#[serde(rename = "ETag")]
+	pub etag: s3_xml::Value,
+}
+
+#[derive(Debug, Serialize, PartialEq)]
+pub struct CopyPartResult {
+	#[serde(serialize_with = "xmlns_tag")]
+	pub xmlns: (),
+	#[serde(rename = "LastModified")]
+	pub last_modified: s3_xml::Value,
+	#[serde(rename = "ETag")]
+	pub etag: s3_xml::Value,
+}
+
+#[cfg(test)]
+mod tests {
+	use super::*;
+	use crate::s3_xml::to_xml_with_header;
+
+	#[test]
+	fn copy_object_result() -> Result<(), Error> {
+		let copy_result = CopyObjectResult {
+			last_modified: s3_xml::Value(msec_to_rfc3339(0)),
+			etag: s3_xml::Value("\"9b2cf535f27731c974343645a3985328\"".to_string()),
+		};
+		assert_eq!(
+			to_xml_with_header(&copy_result)?,
+			"<?xml version=\"1.0\" encoding=\"UTF-8\"?>\
+<CopyObjectResult>\
+    <LastModified>1970-01-01T00:00:00.000Z</LastModified>\
+    <ETag>&quot;9b2cf535f27731c974343645a3985328&quot;</ETag>\
+</CopyObjectResult>\
+			"
+		);
+		Ok(())
+	}
+
+	#[test]
+	fn serialize_copy_part_result() -> Result<(), Error> {
+		let expected_retval = "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\
+<CopyPartResult xmlns=\"http://s3.amazonaws.com/doc/2006-03-01/\">\
+	<LastModified>2011-04-11T20:34:56.000Z</LastModified>\
+	<ETag>&quot;9b2cf535f27731c974343645a3985328&quot;</ETag>\
+</CopyPartResult>";
+		let v = CopyPartResult {
+			xmlns: (),
+			last_modified: s3_xml::Value("2011-04-11T20:34:56.000Z".into()),
+			etag: s3_xml::Value("\"9b2cf535f27731c974343645a3985328\"".into()),
+		};
+		println!("{}", to_xml_with_header(&v)?);
+
+		assert_eq!(to_xml_with_header(&v)?, expected_retval);
+
+		Ok(())
+	}
+}

src/api/s3_cors.rs

@@ -0,0 +1,339 @@
+use quick_xml::de::from_reader;
+use std::sync::Arc;
+
+use http::header::{
+	ACCESS_CONTROL_ALLOW_HEADERS, ACCESS_CONTROL_ALLOW_METHODS, ACCESS_CONTROL_ALLOW_ORIGIN,
+	ACCESS_CONTROL_EXPOSE_HEADERS,
+};
+use hyper::{header::HeaderName, Body, Method, Request, Response, StatusCode};
+
+use serde::{Deserialize, Serialize};
+
+use crate::error::*;
+use crate::s3_xml::{to_xml_with_header, xmlns_tag, IntValue, Value};
+use crate::signature::verify_signed_content;
+
+use garage_model::bucket_table::CorsRule as GarageCorsRule;
+use garage_model::garage::Garage;
+use garage_table::*;
+use garage_util::data::*;
+
+pub async fn handle_get_cors(
+	garage: Arc<Garage>,
+	bucket_id: Uuid,
+) -> Result<Response<Body>, Error> {
+	let bucket = garage
+		.bucket_table
+		.get(&EmptyKey, &bucket_id)
+		.await?
+		.ok_or(Error::NoSuchBucket)?;
+
+	let param = bucket
+		.params()
+		.ok_or_internal_error("Bucket should not be deleted at this point")?;
+
+	if let Some(cors) = param.cors_config.get() {
+		let wc = CorsConfiguration {
+			xmlns: (),
+			cors_rules: cors
+				.iter()
+				.map(CorsRule::from_garage_cors_rule)
+				.collect::<Vec<_>>(),
+		};
+		let xml = to_xml_with_header(&wc)?;
+		Ok(Response::builder()
+			.status(StatusCode::OK)
+			.header(http::header::CONTENT_TYPE, "application/xml")
+			.body(Body::from(xml))?)
+	} else {
+		Ok(Response::builder()
+			.status(StatusCode::NO_CONTENT)
+			.body(Body::empty())?)
+	}
+}
+
+pub async fn handle_delete_cors(
+	garage: Arc<Garage>,
+	bucket_id: Uuid,
+) -> Result<Response<Body>, Error> {
+	let mut bucket = garage
+		.bucket_table
+		.get(&EmptyKey, &bucket_id)
+		.await?
+		.ok_or(Error::NoSuchBucket)?;
+
+	let param = bucket
+		.params_mut()
+		.ok_or_internal_error("Bucket should not be deleted at this point")?;
+
+	param.cors_config.update(None);
+	garage.bucket_table.insert(&bucket).await?;
+
+	Ok(Response::builder()
+		.status(StatusCode::NO_CONTENT)
+		.body(Body::empty())?)
+}
+
+pub async fn handle_put_cors(
+	garage: Arc<Garage>,
+	bucket_id: Uuid,
+	req: Request<Body>,
+	content_sha256: Option<Hash>,
+) -> Result<Response<Body>, Error> {
+	let body = hyper::body::to_bytes(req.into_body()).await?;
+	verify_signed_content(content_sha256, &body[..])?;
+
+	let mut bucket = garage
+		.bucket_table
+		.get(&EmptyKey, &bucket_id)
+		.await?
+		.ok_or(Error::NoSuchBucket)?;
+
+	let param = bucket
+		.params_mut()
+		.ok_or_internal_error("Bucket should not be deleted at this point")?;
+
+	let conf: CorsConfiguration = from_reader(&body as &[u8])?;
+	conf.validate()?;
+
+	param
+		.cors_config
+		.update(Some(conf.into_garage_cors_config()?));
+	garage.bucket_table.insert(&bucket).await?;
+
+	Ok(Response::builder()
+		.status(StatusCode::OK)
+		.body(Body::empty())?)
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+#[serde(rename = "CORSConfiguration")]
+pub struct CorsConfiguration {
+	#[serde(serialize_with = "xmlns_tag", skip_deserializing)]
+	pub xmlns: (),
+	#[serde(rename = "CORSRule")]
+	pub cors_rules: Vec<CorsRule>,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct CorsRule {
+	#[serde(rename = "ID")]
+	pub id: Option<Value>,
+	#[serde(rename = "MaxAgeSeconds")]
+	pub max_age_seconds: Option<IntValue>,
+	#[serde(rename = "AllowedOrigin")]
+	pub allowed_origins: Vec<Value>,
+	#[serde(rename = "AllowedMethod")]
+	pub allowed_methods: Vec<Value>,
+	#[serde(rename = "AllowedHeader", default)]
+	pub allowed_headers: Vec<Value>,
+	#[serde(rename = "ExposeHeader", default)]
+	pub expose_headers: Vec<Value>,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct AllowedMethod {
+	#[serde(rename = "AllowedMethod")]
+	pub allowed_method: Value,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct AllowedHeader {
+	#[serde(rename = "AllowedHeader")]
+	pub allowed_header: Value,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct ExposeHeader {
+	#[serde(rename = "ExposeHeader")]
+	pub expose_header: Value,
+}
+
+impl CorsConfiguration {
+	pub fn validate(&self) -> Result<(), Error> {
+		for r in self.cors_rules.iter() {
+			r.validate()?;
+		}
+		Ok(())
+	}
+
+	pub fn into_garage_cors_config(self) -> Result<Vec<GarageCorsRule>, Error> {
+		Ok(self
+			.cors_rules
+			.iter()
+			.map(CorsRule::to_garage_cors_rule)
+			.collect())
+	}
+}
+
+impl CorsRule {
+	pub fn validate(&self) -> Result<(), Error> {
+		for method in self.allowed_methods.iter() {
+			method
+				.0
+				.parse::<Method>()
+				.ok_or_bad_request("Invalid CORSRule method")?;
+		}
+		for header in self
+			.allowed_headers
+			.iter()
+			.chain(self.expose_headers.iter())
+		{
+			header
+				.0
+				.parse::<HeaderName>()
+				.ok_or_bad_request("Invalid HTTP header name")?;
+		}
+		Ok(())
+	}
+
+	pub fn to_garage_cors_rule(&self) -> GarageCorsRule {
+		let convert_vec =
+			|vval: &[Value]| vval.iter().map(|x| x.0.to_owned()).collect::<Vec<String>>();
+		GarageCorsRule {
+			id: self.id.as_ref().map(|x| x.0.to_owned()),
+			max_age_seconds: self.max_age_seconds.as_ref().map(|x| x.0 as u64),
+			allow_origins: convert_vec(&self.allowed_origins),
+			allow_methods: convert_vec(&self.allowed_methods),
+			allow_headers: convert_vec(&self.allowed_headers),
+			expose_headers: convert_vec(&self.expose_headers),
+		}
+	}
+
+	pub fn from_garage_cors_rule(rule: &GarageCorsRule) -> Self {
+		let convert_vec = |vval: &[String]| {
+			vval.iter()
+				.map(|x| Value(x.clone()))
+				.collect::<Vec<Value>>()
+		};
+		Self {
+			id: rule.id.as_ref().map(|x| Value(x.clone())),
+			max_age_seconds: rule.max_age_seconds.map(|x| IntValue(x as i64)),
+			allowed_origins: convert_vec(&rule.allow_origins),
+			allowed_methods: convert_vec(&rule.allow_methods),
+			allowed_headers: convert_vec(&rule.allow_headers),
+			expose_headers: convert_vec(&rule.expose_headers),
+		}
+	}
+}
+
+pub fn cors_rule_matches<'a, HI, S>(
+	rule: &GarageCorsRule,
+	origin: &'a str,
+	method: &'a str,
+	mut request_headers: HI,
+) -> bool
+where
+	HI: Iterator<Item = S>,
+	S: AsRef<str>,
+{
+	rule.allow_origins.iter().any(|x| x == "*" || x == origin)
+		&& rule.allow_methods.iter().any(|x| x == "*" || x == method)
+		&& request_headers.all(|h| {
+			rule.allow_headers
+				.iter()
+				.any(|x| x == "*" || x == h.as_ref())
+		})
+}
+
+pub fn add_cors_headers(
+	resp: &mut Response<Body>,
+	rule: &GarageCorsRule,
+) -> Result<(), http::header::InvalidHeaderValue> {
+	let h = resp.headers_mut();
+	h.insert(
+		ACCESS_CONTROL_ALLOW_ORIGIN,
+		rule.allow_origins.join(", ").parse()?,
+	);
+	h.insert(
+		ACCESS_CONTROL_ALLOW_METHODS,
+		rule.allow_methods.join(", ").parse()?,
+	);
+	h.insert(
+		ACCESS_CONTROL_ALLOW_HEADERS,
+		rule.allow_headers.join(", ").parse()?,
+	);
+	h.insert(
+		ACCESS_CONTROL_EXPOSE_HEADERS,
+		rule.expose_headers.join(", ").parse()?,
+	);
+	Ok(())
+}
+
+#[cfg(test)]
+mod tests {
+	use super::*;
+
+	use quick_xml::de::from_str;
+
+	#[test]
+	fn test_deserialize() -> Result<(), Error> {
+		let message = r#"<?xml version="1.0" encoding="UTF-8"?>
+<CORSConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/">
+ <CORSRule>
+   <AllowedOrigin>http://www.example.com</AllowedOrigin>
+
+   <AllowedMethod>PUT</AllowedMethod>
+   <AllowedMethod>POST</AllowedMethod>
+   <AllowedMethod>DELETE</AllowedMethod>
+
+   <AllowedHeader>*</AllowedHeader>
+ </CORSRule>
+ <CORSRule>
+   <AllowedOrigin>*</AllowedOrigin>
+   <AllowedMethod>GET</AllowedMethod>
+ </CORSRule>
+ <CORSRule>
+   <ID>qsdfjklm</ID>
+   <MaxAgeSeconds>12345</MaxAgeSeconds>
+   <AllowedOrigin>https://perdu.com</AllowedOrigin>
+
+   <AllowedMethod>GET</AllowedMethod>
+   <AllowedMethod>DELETE</AllowedMethod>
+   <AllowedHeader>*</AllowedHeader>
+   <ExposeHeader>*</ExposeHeader>
+ </CORSRule>
+</CORSConfiguration>"#;
+		let conf: CorsConfiguration = from_str(message).unwrap();
+		let ref_value = CorsConfiguration {
+			xmlns: (),
+			cors_rules: vec![
+				CorsRule {
+					id: None,
+					max_age_seconds: None,
+					allowed_origins: vec!["http://www.example.com".into()],
+					allowed_methods: vec!["PUT".into(), "POST".into(), "DELETE".into()],
+					allowed_headers: vec!["*".into()],
+					expose_headers: vec![],
+				},
+				CorsRule {
+					id: None,
+					max_age_seconds: None,
+					allowed_origins: vec!["*".into()],
+					allowed_methods: vec!["GET".into()],
+					allowed_headers: vec![],
+					expose_headers: vec![],
+				},
+				CorsRule {
+					id: Some("qsdfjklm".into()),
+					max_age_seconds: Some(IntValue(12345)),
+					allowed_origins: vec!["https://perdu.com".into()],
+					allowed_methods: vec!["GET".into(), "DELETE".into()],
+					allowed_headers: vec!["*".into()],
+					expose_headers: vec!["*".into()],
+				},
+			],
+		};
+		assert_eq! {
+			ref_value,
+			conf
+		};
+
+		let message2 = to_xml_with_header(&ref_value)?;
+
+		let cleanup = |c: &str| c.replace(char::is_whitespace, "");
+		assert_eq!(cleanup(message), cleanup(&message2));
+
+		Ok(())
+	}
+}

src/api/s3_delete.rs

@@ -1,6 +1,6 @@
 use std::sync::Arc;
 
-use hyper::{Body, Request, Response};
+use hyper::{Body, Request, Response, StatusCode};
 
 use garage_util::data::*;
 use garage_util::time::*;
@@ -14,14 +14,14 @@ use crate::signature::verify_signed_content;
 
 async fn handle_delete_internal(
 	garage: &Garage,
-	bucket: &str,
+	bucket_id: Uuid,
 	key: &str,
 ) -> Result<(Uuid, Uuid), Error> {
 	let object = garage
 		.object_table
-		.get(&bucket.to_string(), &key.to_string())
+		.get(&bucket_id, &key.to_string())
 		.await?
-		.ok_or(Error::NotFound)?; // No need to delete
+		.ok_or(Error::NoSuchKey)?; // No need to delete
 
 	let interesting_versions = object.versions().iter().filter(|v| {
 		!matches!(
@@ -40,12 +40,12 @@ async fn handle_delete_internal(
 		timestamp = std::cmp::max(timestamp, v.timestamp + 1);
 	}
 
-	let deleted_version = version_to_delete.ok_or(Error::NotFound)?;
+	let deleted_version = version_to_delete.ok_or(Error::NoSuchKey)?;
 
 	let version_uuid = gen_uuid();
 
 	let object = Object::new(
-		bucket.into(),
+		bucket_id,
 		key.into(),
 		vec![ObjectVersion {
 			uuid: version_uuid,
@@ -55,40 +55,42 @@ async fn handle_delete_internal(
 	);
 
 	garage.object_table.insert(&object).await?;
-	return Ok((deleted_version, version_uuid));
+
+	Ok((deleted_version, version_uuid))
 }
 
 pub async fn handle_delete(
 	garage: Arc<Garage>,
-	bucket: &str,
+	bucket_id: Uuid,
 	key: &str,
 ) -> Result<Response<Body>, Error> {
 	let (_deleted_version, delete_marker_version) =
-		handle_delete_internal(&garage, bucket, key).await?;
+		handle_delete_internal(&garage, bucket_id, key).await?;
 
 	Ok(Response::builder()
 		.header("x-amz-version-id", hex::encode(delete_marker_version))
+		.status(StatusCode::NO_CONTENT)
 		.body(Body::from(vec![]))
 		.unwrap())
 }
 
 pub async fn handle_delete_objects(
 	garage: Arc<Garage>,
-	bucket: &str,
+	bucket_id: Uuid,
 	req: Request<Body>,
 	content_sha256: Option<Hash>,
 ) -> Result<Response<Body>, Error> {
 	let body = hyper::body::to_bytes(req.into_body()).await?;
 	verify_signed_content(content_sha256, &body[..])?;
 
-	let cmd_xml = roxmltree::Document::parse(&std::str::from_utf8(&body)?)?;
+	let cmd_xml = roxmltree::Document::parse(std::str::from_utf8(&body)?)?;
 	let cmd = parse_delete_objects_xml(&cmd_xml).ok_or_bad_request("Invalid delete XML query")?;
 
 	let mut ret_deleted = Vec::new();
 	let mut ret_errors = Vec::new();
 
 	for obj in cmd.objects.iter() {
-		match handle_delete_internal(&garage, bucket, &obj.key).await {
+		match handle_delete_internal(&garage, bucket_id, &obj.key).await {
 			Ok((deleted_version, delete_marker_version)) => {
 				if cmd.quiet {
 					continue;

src/api/s3_get.rs

@@ -7,6 +7,7 @@ use hyper::body::Bytes;
 use hyper::{Body, Request, Response, StatusCode};
 
 use garage_table::EmptyKey;
+use garage_util::data::*;
 
 use garage_model::garage::Garage;
 use garage_model::object_table::*;
@@ -84,21 +85,21 @@ fn try_answer_cached(
 pub async fn handle_head(
 	garage: Arc<Garage>,
 	req: &Request<Body>,
-	bucket: &str,
+	bucket_id: Uuid,
 	key: &str,
 ) -> Result<Response<Body>, Error> {
 	let object = garage
 		.object_table
-		.get(&bucket.to_string(), &key.to_string())
+		.get(&bucket_id, &key.to_string())
 		.await?
-		.ok_or(Error::NotFound)?;
+		.ok_or(Error::NoSuchKey)?;
 
 	let version = object
 		.versions()
 		.iter()
 		.rev()
 		.find(|v| v.is_data())
-		.ok_or(Error::NotFound)?;
+		.ok_or(Error::NoSuchKey)?;
 
 	let version_meta = match &version.state {
 		ObjectVersionState::Complete(ObjectVersionData::Inline(meta, _)) => meta,
@@ -106,12 +107,12 @@ pub async fn handle_head(
 		_ => unreachable!(),
 	};
 
-	if let Some(cached) = try_answer_cached(&version, version_meta, req) {
+	if let Some(cached) = try_answer_cached(version, version_meta, req) {
 		return Ok(cached);
 	}
 
 	let body: Body = Body::empty();
-	let response = object_headers(&version, version_meta)
+	let response = object_headers(version, version_meta)
 		.header("Content-Length", format!("{}", version_meta.size))
 		.status(StatusCode::OK)
 		.body(body)
@@ -123,44 +124,45 @@ pub async fn handle_head(
 pub async fn handle_get(
 	garage: Arc<Garage>,
 	req: &Request<Body>,
-	bucket: &str,
+	bucket_id: Uuid,
 	key: &str,
 ) -> Result<Response<Body>, Error> {
 	let object = garage
 		.object_table
-		.get(&bucket.to_string(), &key.to_string())
+		.get(&bucket_id, &key.to_string())
 		.await?
-		.ok_or(Error::NotFound)?;
+		.ok_or(Error::NoSuchKey)?;
 
 	let last_v = object
 		.versions()
 		.iter()
 		.rev()
 		.find(|v| v.is_complete())
-		.ok_or(Error::NotFound)?;
+		.ok_or(Error::NoSuchKey)?;
 
 	let last_v_data = match &last_v.state {
 		ObjectVersionState::Complete(x) => x,
 		_ => unreachable!(),
 	};
 	let last_v_meta = match last_v_data {
-		ObjectVersionData::DeleteMarker => return Err(Error::NotFound),
+		ObjectVersionData::DeleteMarker => return Err(Error::NoSuchKey),
 		ObjectVersionData::Inline(meta, _) => meta,
 		ObjectVersionData::FirstBlock(meta, _) => meta,
 	};
 
-	if let Some(cached) = try_answer_cached(&last_v, last_v_meta, req) {
+	if let Some(cached) = try_answer_cached(last_v, last_v_meta, req) {
 		return Ok(cached);
 	}
 
 	let range = match req.headers().get("range") {
 		Some(range) => {
 			let range_str = range.to_str()?;
-			let mut ranges = http_range::HttpRange::parse(range_str, last_v_meta.size)?;
+			let mut ranges = http_range::HttpRange::parse(range_str, last_v_meta.size)
+				.map_err(|e| (e, last_v_meta.size))?;
 			if ranges.len() > 1 {
-				return Err(Error::BadRequest(
-					"Multiple ranges not supported".to_string(),
-				));
+				// garage does not support multi-range requests yet, so we respond with the entire
+				// object when multiple ranges are requested
+				None
 			} else {
 				ranges.pop()
 			}
@@ -179,7 +181,7 @@ pub async fn handle_get(
 		.await;
 	}
 
-	let resp_builder = object_headers(&last_v, last_v_meta)
+	let resp_builder = object_headers(last_v, last_v_meta)
 		.header("Content-Length", format!("{}", last_v_meta.size))
 		.status(StatusCode::OK);
 
@@ -190,11 +192,11 @@ pub async fn handle_get(
 			Ok(resp_builder.body(body)?)
 		}
 		ObjectVersionData::FirstBlock(_, first_block_hash) => {
-			let read_first_block = garage.block_manager.rpc_get_block(&first_block_hash);
+			let read_first_block = garage.block_manager.rpc_get_block(first_block_hash);
 			let get_next_blocks = garage.version_table.get(&last_v.uuid, &EmptyKey);
 
 			let (first_block, version) = futures::try_join!(read_first_block, get_next_blocks)?;
-			let version = version.ok_or(Error::NotFound)?;
+			let version = version.ok_or(Error::NoSuchKey)?;
 
 			let mut blocks = version
 				.blocks
@@ -235,10 +237,6 @@ async fn handle_get_range(
 	begin: u64,
 	end: u64,
 ) -> Result<Response<Body>, Error> {
-	if end > version_meta.size {
-		return Err(Error::BadRequest("Range not included in file".to_string()));
-	}
-
 	let resp_builder = object_headers(version, version_meta)
 		.header("Content-Length", format!("{}", end - begin))
 		.header(
@@ -263,7 +261,7 @@ async fn handle_get_range(
 			let version = garage.version_table.get(&version.uuid, &EmptyKey).await?;
 			let version = match version {
 				Some(v) => v,
-				None => return Err(Error::NotFound),
+				None => return Err(Error::NoSuchKey),
 			};
 
 			// We will store here the list of blocks that have an intersection with the requested

src/api/s3_put.rs

@@ -24,7 +24,7 @@ use crate::signature::verify_signed_content;
 pub async fn handle_put(
 	garage: Arc<Garage>,
 	req: Request<Body>,
-	bucket: &str,
+	bucket_id: Uuid,
 	key: &str,
 	content_sha256: Option<Hash>,
 ) -> Result<Response<Body>, Error> {
@@ -77,7 +77,7 @@ pub async fn handle_put(
 			)),
 		};
 
-		let object = Object::new(bucket.into(), key.into(), vec![object_version]);
+		let object = Object::new(bucket_id, key.into(), vec![object_version]);
 		garage.object_table.insert(&object).await?;
 
 		return Ok(put_response(version_uuid, data_md5sum_hex));
@@ -90,14 +90,14 @@ pub async fn handle_put(
 		timestamp: version_timestamp,
 		state: ObjectVersionState::Uploading(headers.clone()),
 	};
-	let object = Object::new(bucket.into(), key.into(), vec![object_version.clone()]);
+	let object = Object::new(bucket_id, key.into(), vec![object_version.clone()]);
 	garage.object_table.insert(&object).await?;
 
 	// Initialize corresponding entry in version table
 	// Write this entry now, even with empty block list,
 	// to prevent block_ref entries from being deleted (they can be deleted
 	// if the reference a version that isn't found in the version table)
-	let version = Version::new(version_uuid, bucket.into(), key.into(), false);
+	let version = Version::new(version_uuid, bucket_id, key.into(), false);
 	garage.version_table.insert(&version).await?;
 
 	// Transfer data and verify checksum
@@ -127,7 +127,7 @@ pub async fn handle_put(
 		Err(e) => {
 			// Mark object as aborted, this will free the blocks further down
 			object_version.state = ObjectVersionState::Aborted;
-			let object = Object::new(bucket.into(), key.into(), vec![object_version.clone()]);
+			let object = Object::new(bucket_id, key.into(), vec![object_version.clone()]);
 			garage.object_table.insert(&object).await?;
 			return Err(e);
 		}
@@ -143,7 +143,7 @@ pub async fn handle_put(
 		},
 		first_block_hash,
 	));
-	let object = Object::new(bucket.into(), key.into(), vec![object_version]);
+	let object = Object::new(bucket_id, key.into(), vec![object_version]);
 	garage.object_table.insert(&object).await?;
 
 	Ok(put_response(version_uuid, md5sum_hex))
@@ -193,8 +193,8 @@ async fn read_and_put_blocks(
 
 	let mut next_offset = first_block.len();
 	let mut put_curr_version_block = put_block_meta(
-		&garage,
-		&version,
+		garage,
+		version,
 		part_number,
 		0,
 		first_block_hash,
@@ -213,8 +213,8 @@ async fn read_and_put_blocks(
 			let block_hash = blake2sum(&block[..]);
 			let block_len = block.len();
 			put_curr_version_block = put_block_meta(
-				&garage,
-				&version,
+				garage,
+				version,
 				part_number,
 				next_offset as u64,
 				block_hash,
@@ -315,7 +315,8 @@ pub fn put_response(version_uuid: Uuid, md5sum_hex: String) -> Response<Body> {
 pub async fn handle_create_multipart_upload(
 	garage: Arc<Garage>,
 	req: &Request<Body>,
-	bucket: &str,
+	bucket_name: &str,
+	bucket_id: Uuid,
 	key: &str,
 ) -> Result<Response<Body>, Error> {
 	let version_uuid = gen_uuid();
@@ -327,20 +328,20 @@ pub async fn handle_create_multipart_upload(
 		timestamp: now_msec(),
 		state: ObjectVersionState::Uploading(headers),
 	};
-	let object = Object::new(bucket.to_string(), key.to_string(), vec![object_version]);
+	let object = Object::new(bucket_id, key.to_string(), vec![object_version]);
 	garage.object_table.insert(&object).await?;
 
 	// Insert empty version so that block_ref entries refer to something
 	// (they are inserted concurrently with blocks in the version table, so
 	// there is the possibility that they are inserted before the version table
 	// is created, in which case it is allowed to delete them, e.g. in repair_*)
-	let version = Version::new(version_uuid, bucket.into(), key.into(), false);
+	let version = Version::new(version_uuid, bucket_id, key.into(), false);
 	garage.version_table.insert(&version).await?;
 
 	// Send success response
 	let result = s3_xml::InitiateMultipartUploadResult {
 		xmlns: (),
-		bucket: s3_xml::Value(bucket.to_string()),
+		bucket: s3_xml::Value(bucket_name.to_string()),
 		key: s3_xml::Value(key.to_string()),
 		upload_id: s3_xml::Value(hex::encode(version_uuid)),
 	};
@@ -352,17 +353,12 @@ pub async fn handle_create_multipart_upload(
 pub async fn handle_put_part(
 	garage: Arc<Garage>,
 	req: Request<Body>,
-	bucket: &str,
+	bucket_id: Uuid,
 	key: &str,
-	part_number_str: &str,
+	part_number: u64,
 	upload_id: &str,
 	content_sha256: Option<Hash>,
 ) -> Result<Response<Body>, Error> {
-	// Check parameters
-	let part_number = part_number_str
-		.parse::<u64>()
-		.ok_or_bad_request("Invalid part number")?;
-
 	let version_uuid = decode_upload_id(upload_id)?;
 
 	let content_md5 = match req.headers().get("content-md5") {
@@ -371,27 +367,39 @@ pub async fn handle_put_part(
 	};
 
 	// Read first chuck, and at the same time try to get object to see if it exists
-	let bucket = bucket.to_string();
 	let key = key.to_string();
 	let mut chunker = BodyChunker::new(req.into_body(), garage.config.block_size);
 
-	let (object, first_block) =
-		futures::try_join!(garage.object_table.get(&bucket, &key), chunker.next(),)?;
+	let (object, version, first_block) = futures::try_join!(
+		garage.object_table.get(&bucket_id, &key),
+		garage.version_table.get(&version_uuid, &EmptyKey),
+		chunker.next()
+	)?;
 
 	// Check object is valid and multipart block can be accepted
-	let first_block = first_block.ok_or_else(|| Error::BadRequest("Empty body".to_string()))?;
-	let object = object.ok_or_else(|| Error::BadRequest("Object not found".to_string()))?;
+	let first_block = first_block.ok_or_bad_request("Empty body")?;
+	let object = object.ok_or_bad_request("Object not found")?;
 
 	if !object
 		.versions()
 		.iter()
 		.any(|v| v.uuid == version_uuid && v.is_uploading())
 	{
-		return Err(Error::NotFound);
+		return Err(Error::NoSuchUpload);
+	}
+
+	// Check part hasn't already been uploaded
+	if let Some(v) = version {
+		if v.has_part_number(part_number) {
+			return Err(Error::BadRequest(format!(
+				"Part number {} has already been uploaded",
+				part_number
+			)));
+		}
 	}
 
 	// Copy block to store
-	let version = Version::new(version_uuid, bucket, key, false);
+	let version = Version::new(version_uuid, bucket_id, key, false);
 	let first_block_hash = blake2sum(&first_block[..]);
 	let (_, data_md5sum, data_sha256sum) = read_and_put_blocks(
 		&garage,
@@ -429,7 +437,8 @@ pub async fn handle_put_part(
 pub async fn handle_complete_multipart_upload(
 	garage: Arc<Garage>,
 	req: Request<Body>,
-	bucket: &str,
+	bucket_name: &str,
+	bucket_id: Uuid,
 	key: &str,
 	upload_id: &str,
 	content_sha256: Option<Hash>,
@@ -437,7 +446,7 @@ pub async fn handle_complete_multipart_upload(
 	let body = hyper::body::to_bytes(req.into_body()).await?;
 	verify_signed_content(content_sha256, &body[..])?;
 
-	let body_xml = roxmltree::Document::parse(&std::str::from_utf8(&body)?)?;
+	let body_xml = roxmltree::Document::parse(std::str::from_utf8(&body)?)?;
 	let body_list_of_parts = parse_complete_multpart_upload_body(&body_xml)
 		.ok_or_bad_request("Invalid CompleteMultipartUpload XML")?;
 	debug!(
@@ -447,22 +456,21 @@ pub async fn handle_complete_multipart_upload(
 
 	let version_uuid = decode_upload_id(upload_id)?;
 
-	let bucket = bucket.to_string();
 	let key = key.to_string();
 	let (object, version) = futures::try_join!(
-		garage.object_table.get(&bucket, &key),
+		garage.object_table.get(&bucket_id, &key),
 		garage.version_table.get(&version_uuid, &EmptyKey),
 	)?;
 
-	let object = object.ok_or_else(|| Error::BadRequest("Object not found".to_string()))?;
+	let object = object.ok_or(Error::NoSuchKey)?;
 	let mut object_version = object
 		.versions()
 		.iter()
 		.find(|v| v.uuid == version_uuid && v.is_uploading())
 		.cloned()
-		.ok_or_else(|| Error::BadRequest("Version not found".to_string()))?;
+		.ok_or(Error::NoSuchUpload)?;
 
-	let version = version.ok_or_else(|| Error::BadRequest("Version not found".to_string()))?;
+	let version = version.ok_or(Error::NoSuchKey)?;
 	if version.blocks.is_empty() {
 		return Err(Error::BadRequest("No data was uploaded".to_string()));
 	}
@@ -515,16 +523,16 @@ pub async fn handle_complete_multipart_upload(
 		version.blocks.items()[0].1.hash,
 	));
 
-	let final_object = Object::new(bucket.clone(), key.clone(), vec![object_version]);
+	let final_object = Object::new(bucket_id, key.clone(), vec![object_version]);
 	garage.object_table.insert(&final_object).await?;
 
 	// Send response saying ok we're done
 	let result = s3_xml::CompleteMultipartUploadResult {
 		xmlns: (),
 		location: None,
-		bucket: s3_xml::Value(bucket),
+		bucket: s3_xml::Value(bucket_name.to_string()),
 		key: s3_xml::Value(key),
-		etag: s3_xml::Value(etag),
+		etag: s3_xml::Value(format!("\"{}\"", etag)),
 	};
 	let xml = s3_xml::to_xml_with_header(&result)?;
 
@@ -533,7 +541,7 @@ pub async fn handle_complete_multipart_upload(
 
 pub async fn handle_abort_multipart_upload(
 	garage: Arc<Garage>,
-	bucket: &str,
+	bucket_id: Uuid,
 	key: &str,
 	upload_id: &str,
 ) -> Result<Response<Body>, Error> {
@@ -541,21 +549,21 @@ pub async fn handle_abort_multipart_upload(
 
 	let object = garage
 		.object_table
-		.get(&bucket.to_string(), &key.to_string())
+		.get(&bucket_id, &key.to_string())
 		.await?;
-	let object = object.ok_or_else(|| Error::BadRequest("Object not found".to_string()))?;
+	let object = object.ok_or(Error::NoSuchKey)?;
 
 	let object_version = object
 		.versions()
 		.iter()
 		.find(|v| v.uuid == version_uuid && v.is_uploading());
 	let mut object_version = match object_version {
-		None => return Err(Error::NotFound),
+		None => return Err(Error::NoSuchUpload),
 		Some(x) => x.clone(),
 	};
 
 	object_version.state = ObjectVersionState::Aborted;
-	let final_object = Object::new(bucket.to_string(), key.to_string(), vec![object_version]);
+	let final_object = Object::new(bucket_id, key.to_string(), vec![object_version]);
 	garage.object_table.insert(&final_object).await?;
 
 	Ok(Response::new(Body::from(vec![])))
@@ -615,10 +623,10 @@ pub(crate) fn get_headers(req: &Request<Body>) -> Result<ObjectVersionHeaders, E
 	})
 }
 
-fn decode_upload_id(id: &str) -> Result<Uuid, Error> {
-	let id_bin = hex::decode(id).ok_or_bad_request("Invalid upload ID")?;
+pub fn decode_upload_id(id: &str) -> Result<Uuid, Error> {
+	let id_bin = hex::decode(id).map_err(|_| Error::NoSuchUpload)?;
 	if id_bin.len() != 32 {
-		return None.ok_or_bad_request("Invalid upload ID");
+		return Err(Error::NoSuchUpload);
 	}
 	let mut uuid = [0u8; 32];
 	uuid.copy_from_slice(&id_bin[..]);

src/api/s3_website.rs

@@ -0,0 +1,376 @@
+use quick_xml::de::from_reader;
+use std::sync::Arc;
+
+use hyper::{Body, Request, Response, StatusCode};
+use serde::{Deserialize, Serialize};
+
+use crate::error::*;
+use crate::s3_xml::{to_xml_with_header, xmlns_tag, IntValue, Value};
+use crate::signature::verify_signed_content;
+
+use garage_model::bucket_table::*;
+use garage_model::garage::Garage;
+use garage_table::*;
+use garage_util::data::*;
+
+pub async fn handle_get_website(
+	garage: Arc<Garage>,
+	bucket_id: Uuid,
+) -> Result<Response<Body>, Error> {
+	let bucket = garage
+		.bucket_table
+		.get(&EmptyKey, &bucket_id)
+		.await?
+		.ok_or(Error::NoSuchBucket)?;
+
+	let param = bucket
+		.params()
+		.ok_or_internal_error("Bucket should not be deleted at this point")?;
+
+	if let Some(website) = param.website_config.get() {
+		let wc = WebsiteConfiguration {
+			xmlns: (),
+			error_document: website.error_document.as_ref().map(|v| Key {
+				key: Value(v.to_string()),
+			}),
+			index_document: Some(Suffix {
+				suffix: Value(website.index_document.to_string()),
+			}),
+			redirect_all_requests_to: None,
+			routing_rules: None,
+		};
+		let xml = to_xml_with_header(&wc)?;
+		Ok(Response::builder()
+			.status(StatusCode::OK)
+			.header(http::header::CONTENT_TYPE, "application/xml")
+			.body(Body::from(xml))?)
+	} else {
+		Ok(Response::builder()
+			.status(StatusCode::NO_CONTENT)
+			.body(Body::empty())?)
+	}
+}
+
+pub async fn handle_delete_website(
+	garage: Arc<Garage>,
+	bucket_id: Uuid,
+) -> Result<Response<Body>, Error> {
+	let mut bucket = garage
+		.bucket_table
+		.get(&EmptyKey, &bucket_id)
+		.await?
+		.ok_or(Error::NoSuchBucket)?;
+
+	let param = bucket
+		.params_mut()
+		.ok_or_internal_error("Bucket should not be deleted at this point")?;
+
+	param.website_config.update(None);
+	garage.bucket_table.insert(&bucket).await?;
+
+	Ok(Response::builder()
+		.status(StatusCode::NO_CONTENT)
+		.body(Body::empty())?)
+}
+
+pub async fn handle_put_website(
+	garage: Arc<Garage>,
+	bucket_id: Uuid,
+	req: Request<Body>,
+	content_sha256: Option<Hash>,
+) -> Result<Response<Body>, Error> {
+	let body = hyper::body::to_bytes(req.into_body()).await?;
+	verify_signed_content(content_sha256, &body[..])?;
+
+	let mut bucket = garage
+		.bucket_table
+		.get(&EmptyKey, &bucket_id)
+		.await?
+		.ok_or(Error::NoSuchBucket)?;
+
+	let param = bucket
+		.params_mut()
+		.ok_or_internal_error("Bucket should not be deleted at this point")?;
+
+	let conf: WebsiteConfiguration = from_reader(&body as &[u8])?;
+	conf.validate()?;
+
+	param
+		.website_config
+		.update(Some(conf.into_garage_website_config()?));
+	garage.bucket_table.insert(&bucket).await?;
+
+	Ok(Response::builder()
+		.status(StatusCode::OK)
+		.body(Body::empty())?)
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct WebsiteConfiguration {
+	#[serde(serialize_with = "xmlns_tag", skip_deserializing)]
+	pub xmlns: (),
+	#[serde(rename = "ErrorDocument")]
+	pub error_document: Option<Key>,
+	#[serde(rename = "IndexDocument")]
+	pub index_document: Option<Suffix>,
+	#[serde(rename = "RedirectAllRequestsTo")]
+	pub redirect_all_requests_to: Option<Target>,
+	#[serde(rename = "RoutingRules")]
+	pub routing_rules: Option<Vec<RoutingRule>>,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct RoutingRule {
+	#[serde(rename = "RoutingRule")]
+	pub inner: RoutingRuleInner,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct RoutingRuleInner {
+	#[serde(rename = "Condition")]
+	pub condition: Option<Condition>,
+	#[serde(rename = "Redirect")]
+	pub redirect: Redirect,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct Key {
+	#[serde(rename = "Key")]
+	pub key: Value,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct Suffix {
+	#[serde(rename = "Suffix")]
+	pub suffix: Value,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct Target {
+	#[serde(rename = "HostName")]
+	pub hostname: Value,
+	#[serde(rename = "Protocol")]
+	pub protocol: Option<Value>,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct Condition {
+	#[serde(rename = "HttpErrorCodeReturnedEquals")]
+	pub http_error_code: Option<IntValue>,
+	#[serde(rename = "KeyPrefixEquals")]
+	pub prefix: Option<Value>,
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
+pub struct Redirect {
+	#[serde(rename = "HostName")]
+	pub hostname: Option<Value>,
+	#[serde(rename = "Protocol")]
+	pub protocol: Option<Value>,
+	#[serde(rename = "HttpRedirectCode")]
+	pub http_redirect_code: Option<IntValue>,
+	#[serde(rename = "ReplaceKeyPrefixWith")]
+	pub replace_prefix: Option<Value>,
+	#[serde(rename = "ReplaceKeyWith")]
+	pub replace_full: Option<Value>,
+}
+
+impl WebsiteConfiguration {
+	pub fn validate(&self) -> Result<(), Error> {
+		if self.redirect_all_requests_to.is_some()
+			&& (self.error_document.is_some()
+				|| self.index_document.is_some()
+				|| self.routing_rules.is_some())
+		{
+			return Err(Error::BadRequest(
+				"Bad XML: can't have RedirectAllRequestsTo and other fields".to_owned(),
+			));
+		}
+		if let Some(ref ed) = self.error_document {
+			ed.validate()?;
+		}
+		if let Some(ref id) = self.index_document {
+			id.validate()?;
+		}
+		if let Some(ref rart) = self.redirect_all_requests_to {
+			rart.validate()?;
+		}
+		if let Some(ref rrs) = self.routing_rules {
+			for rr in rrs {
+				rr.inner.validate()?;
+			}
+		}
+
+		Ok(())
+	}
+
+	pub fn into_garage_website_config(self) -> Result<WebsiteConfig, Error> {
+		if self.redirect_all_requests_to.is_some() {
+			Err(Error::NotImplemented(
+				"S3 website redirects are not currently implemented in Garage.".into(),
+			))
+		} else if self.routing_rules.map(|x| !x.is_empty()).unwrap_or(false) {
+			Err(Error::NotImplemented(
+				"S3 routing rules are not currently implemented in Garage.".into(),
+			))
+		} else {
+			Ok(WebsiteConfig {
+				index_document: self
+					.index_document
+					.map(|x| x.suffix.0)
+					.unwrap_or_else(|| "index.html".to_string()),
+				error_document: self.error_document.map(|x| x.key.0),
+			})
+		}
+	}
+}
+
+impl Key {
+	pub fn validate(&self) -> Result<(), Error> {
+		if self.key.0.is_empty() {
+			Err(Error::BadRequest(
+				"Bad XML: error document specified but empty".to_owned(),
+			))
+		} else {
+			Ok(())
+		}
+	}
+}
+
+impl Suffix {
+	pub fn validate(&self) -> Result<(), Error> {
+		if self.suffix.0.is_empty() | self.suffix.0.contains('/') {
+			Err(Error::BadRequest(
+				"Bad XML: index document is empty or contains /".to_owned(),
+			))
+		} else {
+			Ok(())
+		}
+	}
+}
+
+impl Target {
+	pub fn validate(&self) -> Result<(), Error> {
+		if let Some(ref protocol) = self.protocol {
+			if protocol.0 != "http" && protocol.0 != "https" {
+				return Err(Error::BadRequest("Bad XML: invalid protocol".to_owned()));
+			}
+		}
+		Ok(())
+	}
+}
+
+impl RoutingRuleInner {
+	pub fn validate(&self) -> Result<(), Error> {
+		let has_prefix = self
+			.condition
+			.as_ref()
+			.map(|c| c.prefix.as_ref())
+			.flatten()
+			.is_some();
+		self.redirect.validate(has_prefix)
+	}
+}
+
+impl Redirect {
+	pub fn validate(&self, has_prefix: bool) -> Result<(), Error> {
+		if self.replace_prefix.is_some() {
+			if self.replace_full.is_some() {
+				return Err(Error::BadRequest(
+					"Bad XML: both ReplaceKeyPrefixWith and ReplaceKeyWith are set".to_owned(),
+				));
+			}
+			if !has_prefix {
+				return Err(Error::BadRequest(
+					"Bad XML: ReplaceKeyPrefixWith is set, but  KeyPrefixEquals isn't".to_owned(),
+				));
+			}
+		}
+		if let Some(ref protocol) = self.protocol {
+			if protocol.0 != "http" && protocol.0 != "https" {
+				return Err(Error::BadRequest("Bad XML: invalid protocol".to_owned()));
+			}
+		}
+		// TODO there are probably more invalide cases, but which ones?
+		Ok(())
+	}
+}
+
+#[cfg(test)]
+mod tests {
+	use super::*;
+
+	use quick_xml::de::from_str;
+
+	#[test]
+	fn test_deserialize() -> Result<(), Error> {
+		let message = r#"<?xml version="1.0" encoding="UTF-8"?>
+<WebsiteConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/">
+   <ErrorDocument>
+         <Key>my-error-doc</Key>
+   </ErrorDocument>
+   <IndexDocument>
+      <Suffix>my-index</Suffix>
+   </IndexDocument>
+   <RedirectAllRequestsTo>
+      <HostName>garage.tld</HostName>
+      <Protocol>https</Protocol>
+   </RedirectAllRequestsTo>
+   <RoutingRules>
+      <RoutingRule>
+         <Condition>
+            <HttpErrorCodeReturnedEquals>404</HttpErrorCodeReturnedEquals>
+            <KeyPrefixEquals>prefix1</KeyPrefixEquals>
+         </Condition>
+         <Redirect>
+            <HostName>gara.ge</HostName>
+            <Protocol>http</Protocol>
+            <HttpRedirectCode>303</HttpRedirectCode>
+            <ReplaceKeyPrefixWith>prefix2</ReplaceKeyPrefixWith>
+            <ReplaceKeyWith>fullkey</ReplaceKeyWith>
+         </Redirect>
+      </RoutingRule>
+   </RoutingRules>
+</WebsiteConfiguration>"#;
+		let conf: WebsiteConfiguration = from_str(message).unwrap();
+		let ref_value = WebsiteConfiguration {
+			xmlns: (),
+			error_document: Some(Key {
+				key: Value("my-error-doc".to_owned()),
+			}),
+			index_document: Some(Suffix {
+				suffix: Value("my-index".to_owned()),
+			}),
+			redirect_all_requests_to: Some(Target {
+				hostname: Value("garage.tld".to_owned()),
+				protocol: Some(Value("https".to_owned())),
+			}),
+			routing_rules: Some(vec![RoutingRule {
+				inner: RoutingRuleInner {
+					condition: Some(Condition {
+						http_error_code: Some(IntValue(404)),
+						prefix: Some(Value("prefix1".to_owned())),
+					}),
+					redirect: Redirect {
+						hostname: Some(Value("gara.ge".to_owned())),
+						protocol: Some(Value("http".to_owned())),
+						http_redirect_code: Some(IntValue(303)),
+						replace_prefix: Some(Value("prefix2".to_owned())),
+						replace_full: Some(Value("fullkey".to_owned())),
+					},
+				},
+			}]),
+		};
+		assert_eq! {
+			ref_value,
+			conf
+		}
+
+		let message2 = to_xml_with_header(&ref_value)?;
+
+		let cleanup = |c: &str| c.replace(char::is_whitespace, "");
+		assert_eq!(cleanup(message), cleanup(&message2));
+
+		Ok(())
+	}
+}

src/api/s3_xml.rs

@@ -1,5 +1,5 @@
 use quick_xml::se::to_string;
-use serde::{Serialize, Serializer};
+use serde::{Deserialize, Serialize, Serializer};
 
 use crate::Error as ApiError;
 
@@ -9,14 +9,20 @@ pub fn to_xml_with_header<T: Serialize>(x: &T) -> Result<String, ApiError> {
 	Ok(xml)
 }
 
-fn xmlns_tag<S: Serializer>(_v: &(), s: S) -> Result<S::Ok, S::Error> {
+pub fn xmlns_tag<S: Serializer>(_v: &(), s: S) -> Result<S::Ok, S::Error> {
 	s.serialize_str("http://s3.amazonaws.com/doc/2006-03-01/")
 }
 
-#[derive(Debug, Serialize, PartialEq)]
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
 pub struct Value(#[serde(rename = "$value")] pub String);
 
-#[derive(Debug, Serialize, PartialEq)]
+impl From<&str> for Value {
+	fn from(s: &str) -> Value {
+		Value(s.to_string())
+	}
+}
+
+#[derive(Debug, Serialize, Deserialize, PartialEq, Eq, PartialOrd, Ord)]
 pub struct IntValue(#[serde(rename = "$value")] pub i64);
 
 #[derive(Debug, Serialize, PartialEq)]
@@ -107,14 +113,6 @@ pub struct DeleteResult {
 	pub errors: Vec<DeleteError>,
 }
 
-#[derive(Debug, Serialize, PartialEq)]
-pub struct CopyObjectResult {
-	#[serde(rename = "LastModified")]
-	pub last_modified: Value,
-	#[serde(rename = "ETag")]
-	pub etag: Value,
-}
-
 #[derive(Debug, Serialize, PartialEq)]
 pub struct InitiateMultipartUploadResult {
 	#[serde(serialize_with = "xmlns_tag")]
@@ -141,6 +139,60 @@ pub struct CompleteMultipartUploadResult {
 	pub etag: Value,
 }
 
+#[derive(Debug, Serialize, PartialEq)]
+pub struct Initiator {
+	#[serde(rename = "DisplayName")]
+	pub display_name: Value,
+	#[serde(rename = "ID")]
+	pub id: Value,
+}
+
+#[derive(Debug, Serialize, PartialEq)]
+pub struct ListMultipartItem {
+	#[serde(rename = "Initiated")]
+	pub initiated: Value,
+	#[serde(rename = "Initiator")]
+	pub initiator: Initiator,
+	#[serde(rename = "Key")]
+	pub key: Value,
+	#[serde(rename = "UploadId")]
+	pub upload_id: Value,
+	#[serde(rename = "Owner")]
+	pub owner: Owner,
+	#[serde(rename = "StorageClass")]
+	pub storage_class: Value,
+}
+
+#[derive(Debug, Serialize, PartialEq)]
+pub struct ListMultipartUploadsResult {
+	#[serde(serialize_with = "xmlns_tag")]
+	pub xmlns: (),
+	#[serde(rename = "Bucket")]
+	pub bucket: Value,
+	#[serde(rename = "KeyMarker")]
+	pub key_marker: Option<Value>,
+	#[serde(rename = "UploadIdMarker")]
+	pub upload_id_marker: Option<Value>,
+	#[serde(rename = "NextKeyMarker")]
+	pub next_key_marker: Option<Value>,
+	#[serde(rename = "NextUploadIdMarker")]
+	pub next_upload_id_marker: Option<Value>,
+	#[serde(rename = "Prefix")]
+	pub prefix: Value,
+	#[serde(rename = "Delimiter")]
+	pub delimiter: Option<Value>,
+	#[serde(rename = "MaxUploads")]
+	pub max_uploads: IntValue,
+	#[serde(rename = "IsTruncated")]
+	pub is_truncated: Value,
+	#[serde(rename = "Upload")]
+	pub upload: Vec<ListMultipartItem>,
+	#[serde(rename = "CommonPrefixes")]
+	pub common_prefixes: Vec<CommonPrefix>,
+	#[serde(rename = "EncodingType")]
+	pub encoding_type: Option<Value>,
+}
+
 #[derive(Debug, Serialize, PartialEq)]
 pub struct ListBucketItem {
 	#[serde(rename = "Key")]
@@ -372,24 +424,6 @@ mod tests {
 		Ok(())
 	}
 
-	#[test]
-	fn copy_object_result() -> Result<(), ApiError> {
-		let copy_result = CopyObjectResult {
-			last_modified: Value(msec_to_rfc3339(0)),
-			etag: Value("9b2cf535f27731c974343645a3985328".to_string()),
-		};
-		assert_eq!(
-			to_xml_with_header(&copy_result)?,
-			"<?xml version=\"1.0\" encoding=\"UTF-8\"?>\
-<CopyObjectResult>\
-    <LastModified>1970-01-01T00:00:00.000Z</LastModified>\
-    <ETag>9b2cf535f27731c974343645a3985328</ETag>\
-</CopyObjectResult>\
-			"
-		);
-		Ok(())
-	}
-
 	#[test]
 	fn initiate_multipart_upload_result() -> Result<(), ApiError> {
 		let result = InitiateMultipartUploadResult {
@@ -417,7 +451,7 @@ mod tests {
 			location: Some(Value("https://garage.tld/mybucket/a/plop".to_string())),
 			bucket: Value("mybucket".to_string()),
 			key: Value("a/plop".to_string()),
-			etag: Value("3858f62230ac3c915f300c664312c11f-9".to_string()),
+			etag: Value("\"3858f62230ac3c915f300c664312c11f-9\"".to_string()),
 		};
 		assert_eq!(
 			to_xml_with_header(&result)?,
@@ -426,12 +460,64 @@ mod tests {
 	<Location>https://garage.tld/mybucket/a/plop</Location>\
 	<Bucket>mybucket</Bucket>\
 	<Key>a/plop</Key>\
-	<ETag>3858f62230ac3c915f300c664312c11f-9</ETag>\
+	<ETag>&quot;3858f62230ac3c915f300c664312c11f-9&quot;</ETag>\
 </CompleteMultipartUploadResult>"
 		);
 		Ok(())
 	}
 
+	#[test]
+	fn list_multipart_uploads_result() -> Result<(), ApiError> {
+		let result = ListMultipartUploadsResult {
+			xmlns: (),
+			bucket: Value("example-bucket".to_string()),
+			key_marker: None,
+			next_key_marker: None,
+			upload_id_marker: None,
+			encoding_type: None,
+			next_upload_id_marker: None,
+			upload: vec![],
+			delimiter: Some(Value("/".to_string())),
+			prefix: Value("photos/2006/".to_string()),
+			max_uploads: IntValue(1000),
+			is_truncated: Value("false".to_string()),
+			common_prefixes: vec![
+				CommonPrefix {
+					prefix: Value("photos/2006/February/".to_string()),
+				},
+				CommonPrefix {
+					prefix: Value("photos/2006/January/".to_string()),
+				},
+				CommonPrefix {
+					prefix: Value("photos/2006/March/".to_string()),
+				},
+			],
+		};
+
+		assert_eq!(
+			to_xml_with_header(&result)?,
+			"<?xml version=\"1.0\" encoding=\"UTF-8\"?>\
+<ListMultipartUploadsResult xmlns=\"http://s3.amazonaws.com/doc/2006-03-01/\">\
+	<Bucket>example-bucket</Bucket>\
+	<Prefix>photos/2006/</Prefix>\
+	<Delimiter>/</Delimiter>\
+	<MaxUploads>1000</MaxUploads>\
+	<IsTruncated>false</IsTruncated>\
+	<CommonPrefixes>\
+		<Prefix>photos/2006/February/</Prefix>\
+	</CommonPrefixes>\
+	<CommonPrefixes>\
+		<Prefix>photos/2006/January/</Prefix>\
+	</CommonPrefixes>\
+	<CommonPrefixes>\
+		<Prefix>photos/2006/March/</Prefix>\
+	</CommonPrefixes>\
+</ListMultipartUploadsResult>"
+		);
+
+		Ok(())
+	}
+
 	#[test]
 	fn list_objects_v1_1() -> Result<(), ApiError> {
 		let result = ListBucketResult {
@@ -451,7 +537,7 @@ mod tests {
 			contents: vec![ListBucketItem {
 				key: Value("sample.jpg".to_string()),
 				last_modified: Value(msec_to_rfc3339(0)),
-				etag: Value("bf1d737a4d46a19f3bced6905cc8b902".to_string()),
+				etag: Value("\"bf1d737a4d46a19f3bced6905cc8b902\"".to_string()),
 				size: IntValue(142863),
 				storage_class: Value("STANDARD".to_string()),
 			}],
@@ -472,7 +558,7 @@ mod tests {
   <Contents>\
     <Key>sample.jpg</Key>\
     <LastModified>1970-01-01T00:00:00.000Z</LastModified>\
-    <ETag>bf1d737a4d46a19f3bced6905cc8b902</ETag>\
+    <ETag>&quot;bf1d737a4d46a19f3bced6905cc8b902&quot;</ETag>\
     <Size>142863</Size>\
     <StorageClass>STANDARD</StorageClass>\
   </Contents>\
@@ -550,7 +636,7 @@ mod tests {
 			contents: vec![ListBucketItem {
 				key: Value("ExampleObject.txt".to_string()),
 				last_modified: Value(msec_to_rfc3339(0)),
-				etag: Value("599bab3ed2c697f1d26842727561fd94".to_string()),
+				etag: Value("\"599bab3ed2c697f1d26842727561fd94\"".to_string()),
 				size: IntValue(857),
 				storage_class: Value("REDUCED_REDUNDANCY".to_string()),
 			}],
@@ -568,7 +654,7 @@ mod tests {
   <Contents>\
     <Key>ExampleObject.txt</Key>\
     <LastModified>1970-01-01T00:00:00.000Z</LastModified>\
-    <ETag>599bab3ed2c697f1d26842727561fd94</ETag>\
+    <ETag>&quot;599bab3ed2c697f1d26842727561fd94&quot;</ETag>\
     <Size>857</Size>\
     <StorageClass>REDUCED_REDUNDANCY</StorageClass>\
   </Contents>\
@@ -598,7 +684,7 @@ mod tests {
 			contents: vec![ListBucketItem {
 				key: Value("happyfacex.jpg".to_string()),
 				last_modified: Value(msec_to_rfc3339(0)),
-				etag: Value("70ee1738b6b21e2c8a43f3a5ab0eee71".to_string()),
+				etag: Value("\"70ee1738b6b21e2c8a43f3a5ab0eee71\"".to_string()),
 				size: IntValue(1111),
 				storage_class: Value("STANDARD".to_string()),
 			}],
@@ -618,7 +704,7 @@ mod tests {
   <Contents>\
     <Key>happyfacex.jpg</Key>\
     <LastModified>1970-01-01T00:00:00.000Z</LastModified>\
-    <ETag>70ee1738b6b21e2c8a43f3a5ab0eee71</ETag>\
+    <ETag>&quot;70ee1738b6b21e2c8a43f3a5ab0eee71&quot;</ETag>\
     <Size>1111</Size>\
     <StorageClass>STANDARD</StorageClass>\
   </Contents>\

src/api/signature.rs

@@ -64,13 +64,14 @@ pub async fn check_signature(
 		.key_table
 		.get(&EmptyKey, &authorization.key_id)
 		.await?
-		.filter(|k| !k.deleted.get())
+		.filter(|k| !k.state.is_deleted())
 		.ok_or_else(|| Error::Forbidden(format!("No such key: {}", authorization.key_id)))?;
+	let key_p = key.params().unwrap();
 
 	let canonical_request = canonical_request(
 		request.method(),
 		&request.uri().path().to_string(),
-		&canonical_query_string(&request.uri()),
+		&canonical_query_string(request.uri()),
 		&headers,
 		&authorization.signed_headers,
 		&authorization.content_sha256,
@@ -79,7 +80,7 @@ pub async fn check_signature(
 
 	let mut hmac = signing_hmac(
 		&date,
-		&key.secret_key,
+		&key_p.secret_key,
 		&garage.config.s3_api.s3_region,
 		"s3",
 	)
@@ -252,7 +253,7 @@ fn canonical_request(
 		method.as_str(),
 		url_path,
 		canonical_query_string,
-		&canonical_header_string(&headers, signed_headers),
+		&canonical_header_string(headers, signed_headers),
 		"",
 		signed_headers,
 		content_sha256,
@@ -265,10 +266,13 @@ fn canonical_header_string(headers: &HashMap<String, String>, signed_headers: &s
 	let mut items = headers
 		.iter()
 		.filter(|(key, _)| signed_headers_vec.contains(&key.as_str()))
-		.map(|(key, value)| key.to_lowercase() + ":" + value.trim())
 		.collect::<Vec<_>>();
-	items.sort();
-	items.join("\n")
+	items.sort_by(|(k1, _), (k2, _)| k1.cmp(k2));
+	items
+		.iter()
+		.map(|(key, value)| key.to_lowercase() + ":" + value.trim())
+		.collect::<Vec<_>>()
+		.join("\n")
 }
 
 fn canonical_query_string(uri: &hyper::Uri) -> String {

src/garage/Cargo.toml

@@ -1,11 +1,12 @@
 [package]
 name = "garage"
-version = "0.3.0"
+version = "0.6.0"
 authors = ["Alex Auvolat <alex@adnab.me>"]
 edition = "2018"
 license = "AGPL-3.0"
 description = "Garage, an S3-compatible distributed object store for self-hosted deployments"
 repository = "https://git.deuxfleurs.fr/Deuxfleurs/garage"
+readme = "../../README.md"
 
 [[bin]]
 name = "garage"
@@ -14,12 +15,12 @@ path = "main.rs"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-garage_api = { version = "0.3.0", path = "../api" }
-garage_model = { version = "0.3.0", path = "../model" }
-garage_rpc = { version = "0.3.0", path = "../rpc" }
-garage_table = { version = "0.3.0", path = "../table" }
-garage_util = { version = "0.3.0", path = "../util" }
-garage_web = { version = "0.3.0", path = "../web" }
+garage_api = { version = "0.6.0", path = "../api" }
+garage_model = { version = "0.6.0", path = "../model" }
+garage_rpc = { version = "0.6.0", path = "../rpc" }
+garage_table = { version = "0.6.0", path = "../table" }
+garage_util = { version = "0.6.0", path = "../util" }
+garage_web = { version = "0.6.0", path = "../web" }
 
 bytes = "1.0"
 git-version = "0.3.4"
@@ -27,14 +28,20 @@ hex = "0.4"
 log = "0.4"
 pretty_env_logger = "0.4"
 rand = "0.8"
+async-trait = "0.1.7"
+sodiumoxide = { version = "0.2.5-0", package = "kuska-sodiumoxide" }
 
 sled = "0.34"
 
 rmp-serde = "0.15"
 serde = { version = "1.0", default-features = false, features = ["derive", "rc"] }
+serde_bytes = "0.11"
 structopt = { version = "0.3", default-features = false }
 toml = "0.5"
 
 futures = "0.3"
 futures-util = "0.3"
 tokio = { version = "1.0", default-features = false, features = ["rt", "rt-multi-thread", "io-util", "net", "time", "macros", "sync", "signal", "fs"] }
+
+#netapp = { version = "0.3.0", git = "https://git.deuxfleurs.fr/lx/netapp" }
+netapp = "0.3.0"

src/garage/admin.rs

@@ -0,0 +1,776 @@
+use std::collections::HashMap;
+use std::fmt::Write;
+use std::sync::Arc;
+
+use async_trait::async_trait;
+use serde::{Deserialize, Serialize};
+
+use garage_util::crdt::*;
+use garage_util::data::*;
+use garage_util::error::Error as GarageError;
+use garage_util::time::*;
+
+use garage_table::replication::*;
+use garage_table::*;
+
+use garage_rpc::*;
+
+use garage_model::bucket_alias_table::*;
+use garage_model::bucket_table::*;
+use garage_model::garage::Garage;
+use garage_model::helper::error::{Error, OkOrBadRequest};
+use garage_model::key_table::*;
+use garage_model::migrate::Migrate;
+use garage_model::object_table::ObjectFilter;
+use garage_model::permission::*;
+
+use crate::cli::*;
+use crate::repair::Repair;
+
+pub const ADMIN_RPC_PATH: &str = "garage/admin_rpc.rs/Rpc";
+
+#[derive(Debug, Serialize, Deserialize)]
+pub enum AdminRpc {
+	BucketOperation(BucketOperation),
+	KeyOperation(KeyOperation),
+	LaunchRepair(RepairOpt),
+	Migrate(MigrateOpt),
+	Stats(StatsOpt),
+
+	// Replies
+	Ok(String),
+	BucketList(Vec<Bucket>),
+	BucketInfo(Bucket, HashMap<String, Key>),
+	KeyList(Vec<(String, String)>),
+	KeyInfo(Key, HashMap<Uuid, Bucket>),
+}
+
+impl Rpc for AdminRpc {
+	type Response = Result<AdminRpc, Error>;
+}
+
+pub struct AdminRpcHandler {
+	garage: Arc<Garage>,
+	endpoint: Arc<Endpoint<AdminRpc, Self>>,
+}
+
+impl AdminRpcHandler {
+	pub fn new(garage: Arc<Garage>) -> Arc<Self> {
+		let endpoint = garage.system.netapp.endpoint(ADMIN_RPC_PATH.into());
+		let admin = Arc::new(Self { garage, endpoint });
+		admin.endpoint.set_handler(admin.clone());
+		admin
+	}
+
+	async fn handle_bucket_cmd(&self, cmd: &BucketOperation) -> Result<AdminRpc, Error> {
+		match cmd {
+			BucketOperation::List => self.handle_list_buckets().await,
+			BucketOperation::Info(query) => self.handle_bucket_info(query).await,
+			BucketOperation::Create(query) => self.handle_create_bucket(&query.name).await,
+			BucketOperation::Delete(query) => self.handle_delete_bucket(query).await,
+			BucketOperation::Alias(query) => self.handle_alias_bucket(query).await,
+			BucketOperation::Unalias(query) => self.handle_unalias_bucket(query).await,
+			BucketOperation::Allow(query) => self.handle_bucket_allow(query).await,
+			BucketOperation::Deny(query) => self.handle_bucket_deny(query).await,
+			BucketOperation::Website(query) => self.handle_bucket_website(query).await,
+		}
+	}
+
+	async fn handle_list_buckets(&self) -> Result<AdminRpc, Error> {
+		let buckets = self
+			.garage
+			.bucket_table
+			.get_range(&EmptyKey, None, Some(DeletedFilter::NotDeleted), 10000)
+			.await?;
+		Ok(AdminRpc::BucketList(buckets))
+	}
+
+	async fn handle_bucket_info(&self, query: &BucketOpt) -> Result<AdminRpc, Error> {
+		let bucket_id = self
+			.garage
+			.bucket_helper()
+			.resolve_global_bucket_name(&query.name)
+			.await?
+			.ok_or_bad_request("Bucket not found")?;
+
+		let bucket = self
+			.garage
+			.bucket_helper()
+			.get_existing_bucket(bucket_id)
+			.await?;
+
+		let mut relevant_keys = HashMap::new();
+		for (k, _) in bucket
+			.state
+			.as_option()
+			.unwrap()
+			.authorized_keys
+			.items()
+			.iter()
+		{
+			if let Some(key) = self
+				.garage
+				.key_table
+				.get(&EmptyKey, k)
+				.await?
+				.filter(|k| !k.is_deleted())
+			{
+				relevant_keys.insert(k.clone(), key);
+			}
+		}
+		for ((k, _), _, _) in bucket
+			.state
+			.as_option()
+			.unwrap()
+			.local_aliases
+			.items()
+			.iter()
+		{
+			if relevant_keys.contains_key(k) {
+				continue;
+			}
+			if let Some(key) = self.garage.key_table.get(&EmptyKey, k).await? {
+				relevant_keys.insert(k.clone(), key);
+			}
+		}
+
+		Ok(AdminRpc::BucketInfo(bucket, relevant_keys))
+	}
+
+	#[allow(clippy::ptr_arg)]
+	async fn handle_create_bucket(&self, name: &String) -> Result<AdminRpc, Error> {
+		if !is_valid_bucket_name(name) {
+			return Err(Error::BadRequest(format!(
+				"{}: {}",
+				name, INVALID_BUCKET_NAME_MESSAGE
+			)));
+		}
+
+		if let Some(alias) = self.garage.bucket_alias_table.get(&EmptyKey, name).await? {
+			if alias.state.get().is_some() {
+				return Err(Error::BadRequest(format!("Bucket {} already exists", name)));
+			}
+		}
+
+		// ---- done checking, now commit ----
+
+		let bucket = Bucket::new();
+		self.garage.bucket_table.insert(&bucket).await?;
+
+		self.garage
+			.bucket_helper()
+			.set_global_bucket_alias(bucket.id, name)
+			.await?;
+
+		Ok(AdminRpc::Ok(format!("Bucket {} was created.", name)))
+	}
+
+	async fn handle_delete_bucket(&self, query: &DeleteBucketOpt) -> Result<AdminRpc, Error> {
+		let helper = self.garage.bucket_helper();
+
+		let bucket_id = helper
+			.resolve_global_bucket_name(&query.name)
+			.await?
+			.ok_or_bad_request("Bucket not found")?;
+
+		// Get the alias, but keep in minde here the bucket name
+		// given in parameter can also be directly the bucket's ID.
+		// In that case bucket_alias will be None, and
+		// we can still delete the bucket if it has zero aliases
+		// (a condition which we try to prevent but that could still happen somehow).
+		// We just won't try to delete an alias entry because there isn't one.
+		let bucket_alias = self
+			.garage
+			.bucket_alias_table
+			.get(&EmptyKey, &query.name)
+			.await?;
+
+		// Check bucket doesn't have other aliases
+		let mut bucket = helper.get_existing_bucket(bucket_id).await?;
+		let bucket_state = bucket.state.as_option().unwrap();
+		if bucket_state
+			.aliases
+			.items()
+			.iter()
+			.filter(|(_, _, active)| *active)
+			.any(|(name, _, _)| name != &query.name)
+		{
+			return Err(Error::BadRequest(format!("Bucket {} still has other global aliases. Use `bucket unalias` to delete them one by one.", query.name)));
+		}
+		if bucket_state
+			.local_aliases
+			.items()
+			.iter()
+			.any(|(_, _, active)| *active)
+		{
+			return Err(Error::BadRequest(format!("Bucket {} still has other local aliases. Use `bucket unalias` to delete them one by one.", query.name)));
+		}
+
+		// Check bucket is empty
+		let objects = self
+			.garage
+			.object_table
+			.get_range(&bucket_id, None, Some(ObjectFilter::IsData), 10)
+			.await?;
+		if !objects.is_empty() {
+			return Err(Error::BadRequest(format!(
+				"Bucket {} is not empty",
+				query.name
+			)));
+		}
+
+		if !query.yes {
+			return Err(Error::BadRequest(
+				"Add --yes flag to really perform this operation".to_string(),
+			));
+		}
+
+		// --- done checking, now commit ---
+		// 1. delete authorization from keys that had access
+		for (key_id, _) in bucket.authorized_keys() {
+			helper
+				.set_bucket_key_permissions(bucket.id, key_id, BucketKeyPerm::NO_PERMISSIONS)
+				.await?;
+		}
+
+		// 2. delete bucket alias
+		if bucket_alias.is_some() {
+			helper
+				.purge_global_bucket_alias(bucket_id, &query.name)
+				.await?;
+		}
+
+		// 3. delete bucket
+		bucket.state = Deletable::delete();
+		self.garage.bucket_table.insert(&bucket).await?;
+
+		Ok(AdminRpc::Ok(format!("Bucket {} was deleted.", query.name)))
+	}
+
+	async fn handle_alias_bucket(&self, query: &AliasBucketOpt) -> Result<AdminRpc, Error> {
+		let helper = self.garage.bucket_helper();
+
+		let bucket_id = helper
+			.resolve_global_bucket_name(&query.existing_bucket)
+			.await?
+			.ok_or_bad_request("Bucket not found")?;
+
+		if let Some(key_pattern) = &query.local {
+			let key = helper.get_existing_matching_key(key_pattern).await?;
+
+			helper
+				.set_local_bucket_alias(bucket_id, &key.key_id, &query.new_name)
+				.await?;
+			Ok(AdminRpc::Ok(format!(
+				"Alias {} now points to bucket {:?} in namespace of key {}",
+				query.new_name, bucket_id, key.key_id
+			)))
+		} else {
+			helper
+				.set_global_bucket_alias(bucket_id, &query.new_name)
+				.await?;
+			Ok(AdminRpc::Ok(format!(
+				"Alias {} now points to bucket {:?}",
+				query.new_name, bucket_id
+			)))
+		}
+	}
+
+	async fn handle_unalias_bucket(&self, query: &UnaliasBucketOpt) -> Result<AdminRpc, Error> {
+		let helper = self.garage.bucket_helper();
+
+		if let Some(key_pattern) = &query.local {
+			let key = helper.get_existing_matching_key(key_pattern).await?;
+
+			let bucket_id = key
+				.state
+				.as_option()
+				.unwrap()
+				.local_aliases
+				.get(&query.name)
+				.cloned()
+				.flatten()
+				.ok_or_bad_request("Bucket not found")?;
+
+			helper
+				.unset_local_bucket_alias(bucket_id, &key.key_id, &query.name)
+				.await?;
+
+			Ok(AdminRpc::Ok(format!(
+				"Alias {} no longer points to bucket {:?} in namespace of key {}",
+				&query.name, bucket_id, key.key_id
+			)))
+		} else {
+			let bucket_id = helper
+				.resolve_global_bucket_name(&query.name)
+				.await?
+				.ok_or_bad_request("Bucket not found")?;
+
+			helper
+				.unset_global_bucket_alias(bucket_id, &query.name)
+				.await?;
+
+			Ok(AdminRpc::Ok(format!(
+				"Alias {} no longer points to bucket {:?}",
+				&query.name, bucket_id
+			)))
+		}
+	}
+
+	async fn handle_bucket_allow(&self, query: &PermBucketOpt) -> Result<AdminRpc, Error> {
+		let helper = self.garage.bucket_helper();
+
+		let bucket_id = helper
+			.resolve_global_bucket_name(&query.bucket)
+			.await?
+			.ok_or_bad_request("Bucket not found")?;
+		let key = helper.get_existing_matching_key(&query.key_pattern).await?;
+
+		let allow_read = query.read || key.allow_read(&bucket_id);
+		let allow_write = query.write || key.allow_write(&bucket_id);
+		let allow_owner = query.owner || key.allow_owner(&bucket_id);
+
+		helper
+			.set_bucket_key_permissions(
+				bucket_id,
+				&key.key_id,
+				BucketKeyPerm {
+					timestamp: now_msec(),
+					allow_read,
+					allow_write,
+					allow_owner,
+				},
+			)
+			.await?;
+
+		Ok(AdminRpc::Ok(format!(
+			"New permissions for {} on {}: read {}, write {}, owner {}.",
+			&key.key_id, &query.bucket, allow_read, allow_write, allow_owner
+		)))
+	}
+
+	async fn handle_bucket_deny(&self, query: &PermBucketOpt) -> Result<AdminRpc, Error> {
+		let helper = self.garage.bucket_helper();
+
+		let bucket_id = helper
+			.resolve_global_bucket_name(&query.bucket)
+			.await?
+			.ok_or_bad_request("Bucket not found")?;
+		let key = helper.get_existing_matching_key(&query.key_pattern).await?;
+
+		let allow_read = !query.read && key.allow_read(&bucket_id);
+		let allow_write = !query.write && key.allow_write(&bucket_id);
+		let allow_owner = !query.owner && key.allow_owner(&bucket_id);
+
+		helper
+			.set_bucket_key_permissions(
+				bucket_id,
+				&key.key_id,
+				BucketKeyPerm {
+					timestamp: now_msec(),
+					allow_read,
+					allow_write,
+					allow_owner,
+				},
+			)
+			.await?;
+
+		Ok(AdminRpc::Ok(format!(
+			"New permissions for {} on {}: read {}, write {}, owner {}.",
+			&key.key_id, &query.bucket, allow_read, allow_write, allow_owner
+		)))
+	}
+
+	async fn handle_bucket_website(&self, query: &WebsiteOpt) -> Result<AdminRpc, Error> {
+		let bucket_id = self
+			.garage
+			.bucket_helper()
+			.resolve_global_bucket_name(&query.bucket)
+			.await?
+			.ok_or_bad_request("Bucket not found")?;
+
+		let mut bucket = self
+			.garage
+			.bucket_helper()
+			.get_existing_bucket(bucket_id)
+			.await?;
+		let bucket_state = bucket.state.as_option_mut().unwrap();
+
+		if !(query.allow ^ query.deny) {
+			return Err(Error::BadRequest(
+				"You must specify exactly one flag, either --allow or --deny".to_string(),
+			));
+		}
+
+		let website = if query.allow {
+			Some(WebsiteConfig {
+				index_document: query.index_document.clone(),
+				error_document: query.error_document.clone(),
+			})
+		} else {
+			None
+		};
+
+		bucket_state.website_config.update(website);
+		self.garage.bucket_table.insert(&bucket).await?;
+
+		let msg = if query.allow {
+			format!("Website access allowed for {}", &query.bucket)
+		} else {
+			format!("Website access denied for {}", &query.bucket)
+		};
+
+		Ok(AdminRpc::Ok(msg))
+	}
+
+	async fn handle_key_cmd(&self, cmd: &KeyOperation) -> Result<AdminRpc, Error> {
+		match cmd {
+			KeyOperation::List => self.handle_list_keys().await,
+			KeyOperation::Info(query) => self.handle_key_info(query).await,
+			KeyOperation::New(query) => self.handle_create_key(query).await,
+			KeyOperation::Rename(query) => self.handle_rename_key(query).await,
+			KeyOperation::Delete(query) => self.handle_delete_key(query).await,
+			KeyOperation::Allow(query) => self.handle_allow_key(query).await,
+			KeyOperation::Deny(query) => self.handle_deny_key(query).await,
+			KeyOperation::Import(query) => self.handle_import_key(query).await,
+		}
+	}
+
+	async fn handle_list_keys(&self) -> Result<AdminRpc, Error> {
+		let key_ids = self
+			.garage
+			.key_table
+			.get_range(
+				&EmptyKey,
+				None,
+				Some(KeyFilter::Deleted(DeletedFilter::NotDeleted)),
+				10000,
+			)
+			.await?
+			.iter()
+			.map(|k| (k.key_id.to_string(), k.params().unwrap().name.get().clone()))
+			.collect::<Vec<_>>();
+		Ok(AdminRpc::KeyList(key_ids))
+	}
+
+	async fn handle_key_info(&self, query: &KeyOpt) -> Result<AdminRpc, Error> {
+		let key = self
+			.garage
+			.bucket_helper()
+			.get_existing_matching_key(&query.key_pattern)
+			.await?;
+		self.key_info_result(key).await
+	}
+
+	async fn handle_create_key(&self, query: &KeyNewOpt) -> Result<AdminRpc, Error> {
+		let key = Key::new(&query.name);
+		self.garage.key_table.insert(&key).await?;
+		self.key_info_result(key).await
+	}
+
+	async fn handle_rename_key(&self, query: &KeyRenameOpt) -> Result<AdminRpc, Error> {
+		let mut key = self
+			.garage
+			.bucket_helper()
+			.get_existing_matching_key(&query.key_pattern)
+			.await?;
+		key.params_mut()
+			.unwrap()
+			.name
+			.update(query.new_name.clone());
+		self.garage.key_table.insert(&key).await?;
+		self.key_info_result(key).await
+	}
+
+	async fn handle_delete_key(&self, query: &KeyDeleteOpt) -> Result<AdminRpc, Error> {
+		let helper = self.garage.bucket_helper();
+
+		let mut key = helper.get_existing_matching_key(&query.key_pattern).await?;
+
+		if !query.yes {
+			return Err(Error::BadRequest(
+				"Add --yes flag to really perform this operation".to_string(),
+			));
+		}
+
+		let state = key.state.as_option_mut().unwrap();
+
+		// --- done checking, now commit ---
+		// (the step at unset_local_bucket_alias will fail if a bucket
+		// does not have another alias, the deletion will be
+		// interrupted in the middle if that happens)
+
+		// 1. Delete local aliases
+		for (alias, _, to) in state.local_aliases.items().iter() {
+			if let Some(bucket_id) = to {
+				helper
+					.unset_local_bucket_alias(*bucket_id, &key.key_id, alias)
+					.await?;
+			}
+		}
+
+		// 2. Remove permissions on all authorized buckets
+		for (ab_id, _auth) in state.authorized_buckets.items().iter() {
+			helper
+				.set_bucket_key_permissions(*ab_id, &key.key_id, BucketKeyPerm::NO_PERMISSIONS)
+				.await?;
+		}
+
+		// 3. Actually delete key
+		key.state = Deletable::delete();
+		self.garage.key_table.insert(&key).await?;
+
+		Ok(AdminRpc::Ok(format!(
+			"Key {} was deleted successfully.",
+			key.key_id
+		)))
+	}
+
+	async fn handle_allow_key(&self, query: &KeyPermOpt) -> Result<AdminRpc, Error> {
+		let mut key = self
+			.garage
+			.bucket_helper()
+			.get_existing_matching_key(&query.key_pattern)
+			.await?;
+		if query.create_bucket {
+			key.params_mut().unwrap().allow_create_bucket.update(true);
+		}
+		self.garage.key_table.insert(&key).await?;
+		self.key_info_result(key).await
+	}
+
+	async fn handle_deny_key(&self, query: &KeyPermOpt) -> Result<AdminRpc, Error> {
+		let mut key = self
+			.garage
+			.bucket_helper()
+			.get_existing_matching_key(&query.key_pattern)
+			.await?;
+		if query.create_bucket {
+			key.params_mut().unwrap().allow_create_bucket.update(false);
+		}
+		self.garage.key_table.insert(&key).await?;
+		self.key_info_result(key).await
+	}
+
+	async fn handle_import_key(&self, query: &KeyImportOpt) -> Result<AdminRpc, Error> {
+		let prev_key = self.garage.key_table.get(&EmptyKey, &query.key_id).await?;
+		if prev_key.is_some() {
+			return Err(Error::BadRequest(format!("Key {} already exists in data store. Even if it is deleted, we can't let you create a new key with the same ID. Sorry.", query.key_id)));
+		}
+		let imported_key = Key::import(&query.key_id, &query.secret_key, &query.name);
+		self.garage.key_table.insert(&imported_key).await?;
+
+		self.key_info_result(imported_key).await
+	}
+
+	async fn key_info_result(&self, key: Key) -> Result<AdminRpc, Error> {
+		let mut relevant_buckets = HashMap::new();
+
+		for (id, _) in key
+			.state
+			.as_option()
+			.unwrap()
+			.authorized_buckets
+			.items()
+			.iter()
+		{
+			if let Some(b) = self.garage.bucket_table.get(&EmptyKey, id).await? {
+				relevant_buckets.insert(*id, b);
+			}
+		}
+
+		Ok(AdminRpc::KeyInfo(key, relevant_buckets))
+	}
+
+	async fn handle_migrate(self: &Arc<Self>, opt: MigrateOpt) -> Result<AdminRpc, Error> {
+		if !opt.yes {
+			return Err(Error::BadRequest(
+				"Please provide the --yes flag to initiate migration operation.".to_string(),
+			));
+		}
+
+		let m = Migrate {
+			garage: self.garage.clone(),
+		};
+		match opt.what {
+			MigrateWhat::Buckets050 => m.migrate_buckets050().await,
+		}?;
+		Ok(AdminRpc::Ok("Migration successfull.".into()))
+	}
+
+	async fn handle_launch_repair(self: &Arc<Self>, opt: RepairOpt) -> Result<AdminRpc, Error> {
+		if !opt.yes {
+			return Err(Error::BadRequest(
+				"Please provide the --yes flag to initiate repair operations.".to_string(),
+			));
+		}
+		if opt.all_nodes {
+			let mut opt_to_send = opt.clone();
+			opt_to_send.all_nodes = false;
+
+			let mut failures = vec![];
+			let ring = self.garage.system.ring.borrow().clone();
+			for node in ring.layout.node_ids().iter() {
+				let node = (*node).into();
+				let resp = self
+					.endpoint
+					.call(
+						&node,
+						&AdminRpc::LaunchRepair(opt_to_send.clone()),
+						PRIO_NORMAL,
+					)
+					.await;
+				if !matches!(resp, Ok(Ok(_))) {
+					failures.push(node);
+				}
+			}
+			if failures.is_empty() {
+				Ok(AdminRpc::Ok("Repair launched on all nodes".to_string()))
+			} else {
+				Err(Error::BadRequest(format!(
+					"Could not launch repair on nodes: {:?} (launched successfully on other nodes)",
+					failures
+				)))
+			}
+		} else {
+			let repair = Repair {
+				garage: self.garage.clone(),
+			};
+			self.garage
+				.system
+				.background
+				.spawn_worker("Repair worker".into(), move |must_exit| async move {
+					repair.repair_worker(opt, must_exit).await
+				});
+			Ok(AdminRpc::Ok(format!(
+				"Repair launched on {:?}",
+				self.garage.system.id
+			)))
+		}
+	}
+
+	async fn handle_stats(&self, opt: StatsOpt) -> Result<AdminRpc, Error> {
+		if opt.all_nodes {
+			let mut ret = String::new();
+			let ring = self.garage.system.ring.borrow().clone();
+
+			for node in ring.layout.node_ids().iter() {
+				let mut opt = opt.clone();
+				opt.all_nodes = false;
+
+				writeln!(&mut ret, "\n======================").unwrap();
+				writeln!(&mut ret, "Stats for node {:?}:", node).unwrap();
+
+				let node_id = (*node).into();
+				match self
+					.endpoint
+					.call(&node_id, &AdminRpc::Stats(opt), PRIO_NORMAL)
+					.await?
+				{
+					Ok(AdminRpc::Ok(s)) => writeln!(&mut ret, "{}", s).unwrap(),
+					Ok(x) => writeln!(&mut ret, "Bad answer: {:?}", x).unwrap(),
+					Err(e) => writeln!(&mut ret, "Error: {}", e).unwrap(),
+				}
+			}
+			Ok(AdminRpc::Ok(ret))
+		} else {
+			Ok(AdminRpc::Ok(self.gather_stats_local(opt)))
+		}
+	}
+
+	fn gather_stats_local(&self, opt: StatsOpt) -> String {
+		let mut ret = String::new();
+		writeln!(
+			&mut ret,
+			"\nGarage version: {}",
+			option_env!("GIT_VERSION").unwrap_or(git_version::git_version!(
+				prefix = "git:",
+				cargo_prefix = "cargo:",
+				fallback = "unknown"
+			))
+		)
+		.unwrap();
+
+		// Gather ring statistics
+		let ring = self.garage.system.ring.borrow().clone();
+		let mut ring_nodes = HashMap::new();
+		for (_i, loc) in ring.partitions().iter() {
+			for n in ring.get_nodes(loc, ring.replication_factor).iter() {
+				if !ring_nodes.contains_key(n) {
+					ring_nodes.insert(*n, 0usize);
+				}
+				*ring_nodes.get_mut(n).unwrap() += 1;
+			}
+		}
+		writeln!(&mut ret, "\nRing nodes & partition count:").unwrap();
+		for (n, c) in ring_nodes.iter() {
+			writeln!(&mut ret, "  {:?} {}", n, c).unwrap();
+		}
+
+		self.gather_table_stats(&mut ret, &self.garage.bucket_table, &opt);
+		self.gather_table_stats(&mut ret, &self.garage.key_table, &opt);
+		self.gather_table_stats(&mut ret, &self.garage.object_table, &opt);
+		self.gather_table_stats(&mut ret, &self.garage.version_table, &opt);
+		self.gather_table_stats(&mut ret, &self.garage.block_ref_table, &opt);
+
+		writeln!(&mut ret, "\nBlock manager stats:").unwrap();
+		if opt.detailed {
+			writeln!(
+				&mut ret,
+				"  number of RC entries (~= number of blocks): {}",
+				self.garage.block_manager.rc_len()
+			)
+			.unwrap();
+		}
+		writeln!(
+			&mut ret,
+			"  resync queue length: {}",
+			self.garage.block_manager.resync_queue_len()
+		)
+		.unwrap();
+
+		ret
+	}
+
+	fn gather_table_stats<F, R>(&self, to: &mut String, t: &Arc<Table<F, R>>, opt: &StatsOpt)
+	where
+		F: TableSchema + 'static,
+		R: TableReplication + 'static,
+	{
+		writeln!(to, "\nTable stats for {}", F::TABLE_NAME).unwrap();
+		if opt.detailed {
+			writeln!(to, "  number of items: {}", t.data.store.len()).unwrap();
+			writeln!(
+				to,
+				"  Merkle tree size: {}",
+				t.merkle_updater.merkle_tree_len()
+			)
+			.unwrap();
+		}
+		writeln!(
+			to,
+			"  Merkle updater todo queue length: {}",
+			t.merkle_updater.todo_len()
+		)
+		.unwrap();
+		writeln!(to, "  GC todo queue length: {}", t.data.gc_todo_len()).unwrap();
+	}
+}
+
+#[async_trait]
+impl EndpointHandler<AdminRpc> for AdminRpcHandler {
+	async fn handle(
+		self: &Arc<Self>,
+		message: &AdminRpc,
+		_from: NodeID,
+	) -> Result<AdminRpc, Error> {
+		match message {
+			AdminRpc::BucketOperation(bo) => self.handle_bucket_cmd(bo).await,
+			AdminRpc::KeyOperation(ko) => self.handle_key_cmd(ko).await,
+			AdminRpc::Migrate(opt) => self.handle_migrate(opt.clone()).await,
+			AdminRpc::LaunchRepair(opt) => self.handle_launch_repair(opt.clone()).await,
+			AdminRpc::Stats(opt) => self.handle_stats(opt.clone()).await,
+			m => Err(GarageError::unexpected_rpc_message(m).into()),
+		}
+	}
+}

src/garage/admin_rpc.rs

@@ -1,494 +0,0 @@
-use std::collections::HashMap;
-use std::fmt::Write;
-use std::sync::Arc;
-
-use serde::{Deserialize, Serialize};
-
-use garage_util::error::Error;
-
-use garage_table::crdt::Crdt;
-use garage_table::replication::*;
-use garage_table::*;
-
-use garage_rpc::rpc_client::*;
-use garage_rpc::rpc_server::*;
-
-use garage_model::bucket_table::*;
-use garage_model::garage::Garage;
-use garage_model::key_table::*;
-
-use crate::cli::*;
-use crate::repair::Repair;
-use crate::*;
-
-pub const ADMIN_RPC_TIMEOUT: Duration = Duration::from_secs(30);
-pub const ADMIN_RPC_PATH: &str = "_admin";
-
-#[derive(Debug, Serialize, Deserialize)]
-pub enum AdminRpc {
-	BucketOperation(BucketOperation),
-	KeyOperation(KeyOperation),
-	LaunchRepair(RepairOpt),
-	Stats(StatsOpt),
-
-	// Replies
-	Ok(String),
-	BucketList(Vec<String>),
-	BucketInfo(Bucket),
-	KeyList(Vec<(String, String)>),
-	KeyInfo(Key),
-}
-
-impl RpcMessage for AdminRpc {}
-
-pub struct AdminRpcHandler {
-	garage: Arc<Garage>,
-	rpc_client: Arc<RpcClient<AdminRpc>>,
-}
-
-impl AdminRpcHandler {
-	pub fn new(garage: Arc<Garage>) -> Arc<Self> {
-		let rpc_client = garage.system.clone().rpc_client::<AdminRpc>(ADMIN_RPC_PATH);
-		Arc::new(Self { garage, rpc_client })
-	}
-
-	pub fn register_handler(self: Arc<Self>, rpc_server: &mut RpcServer) {
-		rpc_server.add_handler::<AdminRpc, _, _>(ADMIN_RPC_PATH.to_string(), move |msg, _addr| {
-			let self2 = self.clone();
-			async move {
-				match msg {
-					AdminRpc::BucketOperation(bo) => self2.handle_bucket_cmd(bo).await,
-					AdminRpc::KeyOperation(ko) => self2.handle_key_cmd(ko).await,
-					AdminRpc::LaunchRepair(opt) => self2.handle_launch_repair(opt).await,
-					AdminRpc::Stats(opt) => self2.handle_stats(opt).await,
-					_ => Err(Error::BadRpc("Invalid RPC".to_string())),
-				}
-			}
-		});
-	}
-
-	async fn handle_bucket_cmd(&self, cmd: BucketOperation) -> Result<AdminRpc, Error> {
-		match cmd {
-			BucketOperation::List => {
-				let bucket_names = self
-					.garage
-					.bucket_table
-					.get_range(&EmptyKey, None, Some(DeletedFilter::NotDeleted), 10000)
-					.await?
-					.iter()
-					.map(|b| b.name.to_string())
-					.collect::<Vec<_>>();
-				Ok(AdminRpc::BucketList(bucket_names))
-			}
-			BucketOperation::Info(query) => {
-				let bucket = self.get_existing_bucket(&query.name).await?;
-				Ok(AdminRpc::BucketInfo(bucket))
-			}
-			BucketOperation::Create(query) => {
-				let bucket = match self.garage.bucket_table.get(&EmptyKey, &query.name).await? {
-					Some(mut bucket) => {
-						if !bucket.is_deleted() {
-							return Err(Error::BadRpc(format!(
-								"Bucket {} already exists",
-								query.name
-							)));
-						}
-						bucket
-							.state
-							.update(BucketState::Present(BucketParams::new()));
-						bucket
-					}
-					None => Bucket::new(query.name.clone()),
-				};
-				self.garage.bucket_table.insert(&bucket).await?;
-				Ok(AdminRpc::Ok(format!("Bucket {} was created.", query.name)))
-			}
-			BucketOperation::Delete(query) => {
-				let mut bucket = self.get_existing_bucket(&query.name).await?;
-				let objects = self
-					.garage
-					.object_table
-					.get_range(&query.name, None, Some(DeletedFilter::NotDeleted), 10)
-					.await?;
-				if !objects.is_empty() {
-					return Err(Error::BadRpc(format!("Bucket {} is not empty", query.name)));
-				}
-				if !query.yes {
-					return Err(Error::BadRpc(
-						"Add --yes flag to really perform this operation".to_string(),
-					));
-				}
-				// --- done checking, now commit ---
-				for (key_id, _, _) in bucket.authorized_keys() {
-					if let Some(key) = self.garage.key_table.get(&EmptyKey, key_id).await? {
-						if !key.deleted.get() {
-							self.update_key_bucket(&key, &bucket.name, false, false)
-								.await?;
-						}
-					} else {
-						return Err(Error::Message(format!("Key not found: {}", key_id)));
-					}
-				}
-				bucket.state.update(BucketState::Deleted);
-				self.garage.bucket_table.insert(&bucket).await?;
-				Ok(AdminRpc::Ok(format!("Bucket {} was deleted.", query.name)))
-			}
-			BucketOperation::Allow(query) => {
-				let key = self.get_existing_key(&query.key_pattern).await?;
-				let bucket = self.get_existing_bucket(&query.bucket).await?;
-				let allow_read = query.read || key.allow_read(&query.bucket);
-				let allow_write = query.write || key.allow_write(&query.bucket);
-				self.update_key_bucket(&key, &query.bucket, allow_read, allow_write)
-					.await?;
-				self.update_bucket_key(bucket, &key.key_id, allow_read, allow_write)
-					.await?;
-				Ok(AdminRpc::Ok(format!(
-					"New permissions for {} on {}: read {}, write {}.",
-					&key.key_id, &query.bucket, allow_read, allow_write
-				)))
-			}
-			BucketOperation::Deny(query) => {
-				let key = self.get_existing_key(&query.key_pattern).await?;
-				let bucket = self.get_existing_bucket(&query.bucket).await?;
-				let allow_read = !query.read && key.allow_read(&query.bucket);
-				let allow_write = !query.write && key.allow_write(&query.bucket);
-				self.update_key_bucket(&key, &query.bucket, allow_read, allow_write)
-					.await?;
-				self.update_bucket_key(bucket, &key.key_id, allow_read, allow_write)
-					.await?;
-				Ok(AdminRpc::Ok(format!(
-					"New permissions for {} on {}: read {}, write {}.",
-					&key.key_id, &query.bucket, allow_read, allow_write
-				)))
-			}
-			BucketOperation::Website(query) => {
-				let mut bucket = self.get_existing_bucket(&query.bucket).await?;
-
-				if !(query.allow ^ query.deny) {
-					return Err(Error::Message(
-						"You must specify exactly one flag, either --allow or --deny".to_string(),
-					));
-				}
-
-				if let BucketState::Present(state) = bucket.state.get_mut() {
-					state.website.update(query.allow);
-					self.garage.bucket_table.insert(&bucket).await?;
-					let msg = if query.allow {
-						format!("Website access allowed for {}", &query.bucket)
-					} else {
-						format!("Website access denied for {}", &query.bucket)
-					};
-
-					Ok(AdminRpc::Ok(msg))
-				} else {
-					unreachable!();
-				}
-			}
-		}
-	}
-
-	async fn handle_key_cmd(&self, cmd: KeyOperation) -> Result<AdminRpc, Error> {
-		match cmd {
-			KeyOperation::List => {
-				let key_ids = self
-					.garage
-					.key_table
-					.get_range(
-						&EmptyKey,
-						None,
-						Some(KeyFilter::Deleted(DeletedFilter::NotDeleted)),
-						10000,
-					)
-					.await?
-					.iter()
-					.map(|k| (k.key_id.to_string(), k.name.get().clone()))
-					.collect::<Vec<_>>();
-				Ok(AdminRpc::KeyList(key_ids))
-			}
-			KeyOperation::Info(query) => {
-				let key = self.get_existing_key(&query.key_pattern).await?;
-				Ok(AdminRpc::KeyInfo(key))
-			}
-			KeyOperation::New(query) => {
-				let key = Key::new(query.name);
-				self.garage.key_table.insert(&key).await?;
-				Ok(AdminRpc::KeyInfo(key))
-			}
-			KeyOperation::Rename(query) => {
-				let mut key = self.get_existing_key(&query.key_pattern).await?;
-				key.name.update(query.new_name);
-				self.garage.key_table.insert(&key).await?;
-				Ok(AdminRpc::KeyInfo(key))
-			}
-			KeyOperation::Delete(query) => {
-				let key = self.get_existing_key(&query.key_pattern).await?;
-				if !query.yes {
-					return Err(Error::BadRpc(
-						"Add --yes flag to really perform this operation".to_string(),
-					));
-				}
-				// --- done checking, now commit ---
-				for (ab_name, _, _) in key.authorized_buckets.items().iter() {
-					if let Some(bucket) = self.garage.bucket_table.get(&EmptyKey, ab_name).await? {
-						if !bucket.is_deleted() {
-							self.update_bucket_key(bucket, &key.key_id, false, false)
-								.await?;
-						}
-					} else {
-						return Err(Error::Message(format!("Bucket not found: {}", ab_name)));
-					}
-				}
-				let del_key = Key::delete(key.key_id.to_string());
-				self.garage.key_table.insert(&del_key).await?;
-				Ok(AdminRpc::Ok(format!(
-					"Key {} was deleted successfully.",
-					key.key_id
-				)))
-			}
-			KeyOperation::Import(query) => {
-				let prev_key = self.garage.key_table.get(&EmptyKey, &query.key_id).await?;
-				if prev_key.is_some() {
-					return Err(Error::Message(format!("Key {} already exists in data store. Even if it is deleted, we can't let you create a new key with the same ID. Sorry.", query.key_id)));
-				}
-				let imported_key = Key::import(&query.key_id, &query.secret_key, &query.name);
-				self.garage.key_table.insert(&imported_key).await?;
-				Ok(AdminRpc::KeyInfo(imported_key))
-			}
-		}
-	}
-
-	#[allow(clippy::ptr_arg)]
-	async fn get_existing_bucket(&self, bucket: &String) -> Result<Bucket, Error> {
-		self.garage
-			.bucket_table
-			.get(&EmptyKey, bucket)
-			.await?
-			.filter(|b| !b.is_deleted())
-			.map(Ok)
-			.unwrap_or_else(|| Err(Error::BadRpc(format!("Bucket {} does not exist", bucket))))
-	}
-
-	async fn get_existing_key(&self, pattern: &str) -> Result<Key, Error> {
-		let candidates = self
-			.garage
-			.key_table
-			.get_range(
-				&EmptyKey,
-				None,
-				Some(KeyFilter::Matches(pattern.to_string())),
-				10,
-			)
-			.await?
-			.into_iter()
-			.filter(|k| !k.deleted.get())
-			.collect::<Vec<_>>();
-		if candidates.len() != 1 {
-			Err(Error::Message(format!(
-				"{} matching keys",
-				candidates.len()
-			)))
-		} else {
-			Ok(candidates.into_iter().next().unwrap())
-		}
-	}
-
-	/// Update **bucket table** to inform of the new linked key
-	async fn update_bucket_key(
-		&self,
-		mut bucket: Bucket,
-		key_id: &str,
-		allow_read: bool,
-		allow_write: bool,
-	) -> Result<(), Error> {
-		if let BucketState::Present(params) = bucket.state.get_mut() {
-			let ak = &mut params.authorized_keys;
-			let old_ak = ak.take_and_clear();
-			ak.merge(&old_ak.update_mutator(
-				key_id.to_string(),
-				PermissionSet {
-					allow_read,
-					allow_write,
-				},
-			));
-		} else {
-			return Err(Error::Message(
-				"Bucket is deleted in update_bucket_key".to_string(),
-			));
-		}
-		self.garage.bucket_table.insert(&bucket).await?;
-		Ok(())
-	}
-
-	/// Update **key table** to inform of the new linked bucket
-	async fn update_key_bucket(
-		&self,
-		key: &Key,
-		bucket: &str,
-		allow_read: bool,
-		allow_write: bool,
-	) -> Result<(), Error> {
-		let mut key = key.clone();
-		let old_map = key.authorized_buckets.take_and_clear();
-		key.authorized_buckets.merge(&old_map.update_mutator(
-			bucket.to_string(),
-			PermissionSet {
-				allow_read,
-				allow_write,
-			},
-		));
-		self.garage.key_table.insert(&key).await?;
-		Ok(())
-	}
-
-	async fn handle_launch_repair(self: &Arc<Self>, opt: RepairOpt) -> Result<AdminRpc, Error> {
-		if !opt.yes {
-			return Err(Error::BadRpc(
-				"Please provide the --yes flag to initiate repair operations.".to_string(),
-			));
-		}
-		if opt.all_nodes {
-			let mut opt_to_send = opt.clone();
-			opt_to_send.all_nodes = false;
-
-			let mut failures = vec![];
-			let ring = self.garage.system.ring.borrow().clone();
-			for node in ring.config.members.keys() {
-				if self
-					.rpc_client
-					.call(
-						*node,
-						AdminRpc::LaunchRepair(opt_to_send.clone()),
-						ADMIN_RPC_TIMEOUT,
-					)
-					.await
-					.is_err()
-				{
-					failures.push(*node);
-				}
-			}
-			if failures.is_empty() {
-				Ok(AdminRpc::Ok("Repair launched on all nodes".to_string()))
-			} else {
-				Err(Error::Message(format!(
-					"Could not launch repair on nodes: {:?} (launched successfully on other nodes)",
-					failures
-				)))
-			}
-		} else {
-			let repair = Repair {
-				garage: self.garage.clone(),
-			};
-			self.garage
-				.system
-				.background
-				.spawn_worker("Repair worker".into(), move |must_exit| async move {
-					repair.repair_worker(opt, must_exit).await
-				});
-			Ok(AdminRpc::Ok(format!(
-				"Repair launched on {:?}",
-				self.garage.system.id
-			)))
-		}
-	}
-
-	async fn handle_stats(&self, opt: StatsOpt) -> Result<AdminRpc, Error> {
-		if opt.all_nodes {
-			let mut ret = String::new();
-			let ring = self.garage.system.ring.borrow().clone();
-
-			for node in ring.config.members.keys() {
-				let mut opt = opt.clone();
-				opt.all_nodes = false;
-
-				writeln!(&mut ret, "\n======================").unwrap();
-				writeln!(&mut ret, "Stats for node {:?}:", node).unwrap();
-				match self
-					.rpc_client
-					.call(*node, AdminRpc::Stats(opt), ADMIN_RPC_TIMEOUT)
-					.await
-				{
-					Ok(AdminRpc::Ok(s)) => writeln!(&mut ret, "{}", s).unwrap(),
-					Ok(x) => writeln!(&mut ret, "Bad answer: {:?}", x).unwrap(),
-					Err(e) => writeln!(&mut ret, "Error: {}", e).unwrap(),
-				}
-			}
-			Ok(AdminRpc::Ok(ret))
-		} else {
-			Ok(AdminRpc::Ok(self.gather_stats_local(opt)))
-		}
-	}
-
-	fn gather_stats_local(&self, opt: StatsOpt) -> String {
-		let mut ret = String::new();
-		writeln!(
-			&mut ret,
-			"\nGarage version: {}",
-			git_version::git_version!()
-		)
-		.unwrap();
-
-		// Gather ring statistics
-		let ring = self.garage.system.ring.borrow().clone();
-		let mut ring_nodes = HashMap::new();
-		for (_i, loc) in ring.partitions().iter() {
-			for n in ring.get_nodes(loc, ring.replication_factor).iter() {
-				if !ring_nodes.contains_key(n) {
-					ring_nodes.insert(*n, 0usize);
-				}
-				*ring_nodes.get_mut(n).unwrap() += 1;
-			}
-		}
-		writeln!(&mut ret, "\nRing nodes & partition count:").unwrap();
-		for (n, c) in ring_nodes.iter() {
-			writeln!(&mut ret, "  {:?} {}", n, c).unwrap();
-		}
-
-		self.gather_table_stats(&mut ret, &self.garage.bucket_table, &opt);
-		self.gather_table_stats(&mut ret, &self.garage.key_table, &opt);
-		self.gather_table_stats(&mut ret, &self.garage.object_table, &opt);
-		self.gather_table_stats(&mut ret, &self.garage.version_table, &opt);
-		self.gather_table_stats(&mut ret, &self.garage.block_ref_table, &opt);
-
-		writeln!(&mut ret, "\nBlock manager stats:").unwrap();
-		if opt.detailed {
-			writeln!(
-				&mut ret,
-				"  number of blocks: {}",
-				self.garage.block_manager.rc_len()
-			)
-			.unwrap();
-		}
-		writeln!(
-			&mut ret,
-			"  resync queue length: {}",
-			self.garage.block_manager.resync_queue_len()
-		)
-		.unwrap();
-
-		ret
-	}
-
-	fn gather_table_stats<F, R>(&self, to: &mut String, t: &Arc<Table<F, R>>, opt: &StatsOpt)
-	where
-		F: TableSchema + 'static,
-		R: TableReplication + 'static,
-	{
-		writeln!(to, "\nTable stats for {}", t.data.name).unwrap();
-		if opt.detailed {
-			writeln!(to, "  number of items: {}", t.data.store.len()).unwrap();
-			writeln!(
-				to,
-				"  Merkle tree size: {}",
-				t.merkle_updater.merkle_tree_len()
-			)
-			.unwrap();
-		}
-		writeln!(
-			to,
-			"  Merkle updater todo queue length: {}",
-			t.merkle_updater.todo_len()
-		)
-		.unwrap();
-		writeln!(to, "  GC todo queue length: {}", t.data.gc_todo_len()).unwrap();
-	}
-}

src/garage/cli.rs

@@ -1,634 +0,0 @@
-use std::cmp::max;
-use std::collections::HashSet;
-use std::net::SocketAddr;
-use std::path::PathBuf;
-
-use serde::{Deserialize, Serialize};
-use structopt::StructOpt;
-
-use garage_util::data::Uuid;
-use garage_util::error::Error;
-use garage_util::time::*;
-
-use garage_rpc::membership::*;
-use garage_rpc::ring::*;
-use garage_rpc::rpc_client::*;
-
-use garage_model::bucket_table::*;
-use garage_model::key_table::*;
-
-use crate::admin_rpc::*;
-
-#[derive(StructOpt, Debug)]
-pub enum Command {
-	/// Run Garage server
-	#[structopt(name = "server")]
-	Server(ServerOpt),
-
-	/// Get network status
-	#[structopt(name = "status")]
-	Status,
-
-	/// Garage node operations
-	#[structopt(name = "node")]
-	Node(NodeOperation),
-
-	/// Bucket operations
-	#[structopt(name = "bucket")]
-	Bucket(BucketOperation),
-
-	/// Key operations
-	#[structopt(name = "key")]
-	Key(KeyOperation),
-
-	/// Start repair of node data
-	#[structopt(name = "repair")]
-	Repair(RepairOpt),
-
-	/// Gather node statistics
-	#[structopt(name = "stats")]
-	Stats(StatsOpt),
-}
-
-#[derive(StructOpt, Debug)]
-pub struct ServerOpt {
-	/// Configuration file
-	#[structopt(short = "c", long = "config", default_value = "./config.toml")]
-	pub config_file: PathBuf,
-}
-
-#[derive(StructOpt, Debug)]
-pub enum NodeOperation {
-	/// Configure Garage node
-	#[structopt(name = "configure")]
-	Configure(ConfigureNodeOpt),
-
-	/// Remove Garage node from cluster
-	#[structopt(name = "remove")]
-	Remove(RemoveNodeOpt),
-}
-
-#[derive(StructOpt, Debug)]
-pub struct ConfigureNodeOpt {
-	/// Node to configure (prefix of hexadecimal node id)
-	node_id: String,
-
-	/// Location (zone or datacenter) of the node
-	#[structopt(short = "z", long = "zone")]
-	zone: Option<String>,
-
-	/// Capacity (in relative terms, use 1 to represent your smallest server)
-	#[structopt(short = "c", long = "capacity")]
-	capacity: Option<u32>,
-
-	/// Gateway-only node
-	#[structopt(short = "g", long = "gateway")]
-	gateway: bool,
-
-	/// Optional node tag
-	#[structopt(short = "t", long = "tag")]
-	tag: Option<String>,
-
-	/// Replaced node(s): list of node IDs that will be removed from the current cluster
-	#[structopt(long = "replace")]
-	replace: Vec<String>,
-}
-
-#[derive(StructOpt, Debug)]
-pub struct RemoveNodeOpt {
-	/// Node to configure (prefix of hexadecimal node id)
-	node_id: String,
-
-	/// If this flag is not given, the node won't be removed
-	#[structopt(long = "yes")]
-	yes: bool,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub enum BucketOperation {
-	/// List buckets
-	#[structopt(name = "list")]
-	List,
-
-	/// Get bucket info
-	#[structopt(name = "info")]
-	Info(BucketOpt),
-
-	/// Create bucket
-	#[structopt(name = "create")]
-	Create(BucketOpt),
-
-	/// Delete bucket
-	#[structopt(name = "delete")]
-	Delete(DeleteBucketOpt),
-
-	/// Allow key to read or write to bucket
-	#[structopt(name = "allow")]
-	Allow(PermBucketOpt),
-
-	/// Deny key from reading or writing to bucket
-	#[structopt(name = "deny")]
-	Deny(PermBucketOpt),
-
-	/// Expose as website or not
-	#[structopt(name = "website")]
-	Website(WebsiteOpt),
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub struct WebsiteOpt {
-	/// Create
-	#[structopt(long = "allow")]
-	pub allow: bool,
-
-	/// Delete
-	#[structopt(long = "deny")]
-	pub deny: bool,
-
-	/// Bucket name
-	pub bucket: String,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub struct BucketOpt {
-	/// Bucket name
-	pub name: String,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub struct DeleteBucketOpt {
-	/// Bucket name
-	pub name: String,
-
-	/// If this flag is not given, the bucket won't be deleted
-	#[structopt(long = "yes")]
-	pub yes: bool,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub struct PermBucketOpt {
-	/// Access key name or ID
-	#[structopt(long = "key")]
-	pub key_pattern: String,
-
-	/// Allow/deny read operations
-	#[structopt(long = "read")]
-	pub read: bool,
-
-	/// Allow/deny write operations
-	#[structopt(long = "write")]
-	pub write: bool,
-
-	/// Bucket name
-	pub bucket: String,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub enum KeyOperation {
-	/// List keys
-	#[structopt(name = "list")]
-	List,
-
-	/// Get key info
-	#[structopt(name = "info")]
-	Info(KeyOpt),
-
-	/// Create new key
-	#[structopt(name = "new")]
-	New(KeyNewOpt),
-
-	/// Rename key
-	#[structopt(name = "rename")]
-	Rename(KeyRenameOpt),
-
-	/// Delete key
-	#[structopt(name = "delete")]
-	Delete(KeyDeleteOpt),
-
-	/// Import key
-	#[structopt(name = "import")]
-	Import(KeyImportOpt),
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub struct KeyOpt {
-	/// ID or name of the key
-	pub key_pattern: String,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub struct KeyNewOpt {
-	/// Name of the key
-	#[structopt(long = "name", default_value = "Unnamed key")]
-	pub name: String,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub struct KeyRenameOpt {
-	/// ID or name of the key
-	pub key_pattern: String,
-
-	/// New name of the key
-	pub new_name: String,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub struct KeyDeleteOpt {
-	/// ID or name of the key
-	pub key_pattern: String,
-
-	/// Confirm deletion
-	#[structopt(long = "yes")]
-	pub yes: bool,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug)]
-pub struct KeyImportOpt {
-	/// Access key ID
-	pub key_id: String,
-
-	/// Secret access key
-	pub secret_key: String,
-
-	/// Key name
-	#[structopt(short = "n", default_value = "Imported key")]
-	pub name: String,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug, Clone)]
-pub struct RepairOpt {
-	/// Launch repair operation on all nodes
-	#[structopt(short = "a", long = "all-nodes")]
-	pub all_nodes: bool,
-
-	/// Confirm the launch of the repair operation
-	#[structopt(long = "yes")]
-	pub yes: bool,
-
-	#[structopt(subcommand)]
-	pub what: Option<RepairWhat>,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug, Eq, PartialEq, Clone)]
-pub enum RepairWhat {
-	/// Only do a full sync of metadata tables
-	#[structopt(name = "tables")]
-	Tables,
-	/// Only repair (resync/rebalance) the set of stored blocks
-	#[structopt(name = "blocks")]
-	Blocks,
-	/// Only redo the propagation of object deletions to the version table (slow)
-	#[structopt(name = "versions")]
-	Versions,
-	/// Only redo the propagation of version deletions to the block ref table (extremely slow)
-	#[structopt(name = "block_refs")]
-	BlockRefs,
-}
-
-#[derive(Serialize, Deserialize, StructOpt, Debug, Clone)]
-pub struct StatsOpt {
-	/// Gather statistics from all nodes
-	#[structopt(short = "a", long = "all-nodes")]
-	pub all_nodes: bool,
-
-	/// Gather detailed statistics (this can be long)
-	#[structopt(short = "d", long = "detailed")]
-	pub detailed: bool,
-}
-
-pub async fn cli_cmd(
-	cmd: Command,
-	membership_rpc_cli: RpcAddrClient<Message>,
-	admin_rpc_cli: RpcAddrClient<AdminRpc>,
-	rpc_host: SocketAddr,
-) -> Result<(), Error> {
-	match cmd {
-		Command::Status => cmd_status(membership_rpc_cli, rpc_host).await,
-		Command::Node(NodeOperation::Configure(configure_opt)) => {
-			cmd_configure(membership_rpc_cli, rpc_host, configure_opt).await
-		}
-		Command::Node(NodeOperation::Remove(remove_opt)) => {
-			cmd_remove(membership_rpc_cli, rpc_host, remove_opt).await
-		}
-		Command::Bucket(bo) => {
-			cmd_admin(admin_rpc_cli, rpc_host, AdminRpc::BucketOperation(bo)).await
-		}
-		Command::Key(ko) => cmd_admin(admin_rpc_cli, rpc_host, AdminRpc::KeyOperation(ko)).await,
-		Command::Repair(ro) => cmd_admin(admin_rpc_cli, rpc_host, AdminRpc::LaunchRepair(ro)).await,
-		Command::Stats(so) => cmd_admin(admin_rpc_cli, rpc_host, AdminRpc::Stats(so)).await,
-		_ => unreachable!(),
-	}
-}
-
-pub async fn cmd_status(
-	rpc_cli: RpcAddrClient<Message>,
-	rpc_host: SocketAddr,
-) -> Result<(), Error> {
-	let status = match rpc_cli
-		.call(&rpc_host, &Message::PullStatus, ADMIN_RPC_TIMEOUT)
-		.await??
-	{
-		Message::AdvertiseNodesUp(nodes) => nodes,
-		resp => return Err(Error::Message(format!("Invalid RPC response: {:?}", resp))),
-	};
-	let config = match rpc_cli
-		.call(&rpc_host, &Message::PullConfig, ADMIN_RPC_TIMEOUT)
-		.await??
-	{
-		Message::AdvertiseConfig(cfg) => cfg,
-		resp => return Err(Error::Message(format!("Invalid RPC response: {:?}", resp))),
-	};
-
-	let (hostname_len, addr_len, tag_len, zone_len) = status
-		.iter()
-		.map(|adv| (adv, config.members.get(&adv.id)))
-		.map(|(adv, cfg)| {
-			(
-				adv.state_info.hostname.len(),
-				adv.addr.to_string().len(),
-				cfg.map(|c| c.tag.len()).unwrap_or(0),
-				cfg.map(|c| c.zone.len()).unwrap_or(0),
-			)
-		})
-		.fold((0, 0, 0, 0), |(h, a, t, z), (mh, ma, mt, mz)| {
-			(max(h, mh), max(a, ma), max(t, mt), max(z, mz))
-		});
-
-	println!("Healthy nodes:");
-	for adv in status.iter().filter(|x| x.is_up) {
-		if let Some(cfg) = config.members.get(&adv.id) {
-			println!(
-				"{id:?}\t{host}{h_pad}\t{addr}{a_pad}\t[{tag}]{t_pad}\t{zone}{z_pad}\t{capacity}",
-				id = adv.id,
-				host = adv.state_info.hostname,
-				addr = adv.addr,
-				tag = cfg.tag,
-				zone = cfg.zone,
-				capacity = cfg.capacity_string(),
-				h_pad = " ".repeat(hostname_len - adv.state_info.hostname.len()),
-				a_pad = " ".repeat(addr_len - adv.addr.to_string().len()),
-				t_pad = " ".repeat(tag_len - cfg.tag.len()),
-				z_pad = " ".repeat(zone_len - cfg.zone.len()),
-			);
-		} else {
-			println!(
-				"{id:?}\t{h}{h_pad}\t{addr}{a_pad}\tUNCONFIGURED/REMOVED",
-				id = adv.id,
-				h = adv.state_info.hostname,
-				addr = adv.addr,
-				h_pad = " ".repeat(hostname_len - adv.state_info.hostname.len()),
-				a_pad = " ".repeat(addr_len - adv.addr.to_string().len()),
-			);
-		}
-	}
-
-	let status_keys = status.iter().map(|x| x.id).collect::<HashSet<_>>();
-	let failure_case_1 = status.iter().any(|x| !x.is_up);
-	let failure_case_2 = config
-		.members
-		.iter()
-		.any(|(id, _)| !status_keys.contains(id));
-	if failure_case_1 || failure_case_2 {
-		println!("\nFailed nodes:");
-		for adv in status.iter().filter(|x| !x.is_up) {
-			if let Some(cfg) = config.members.get(&adv.id) {
-				println!(
-					"{id:?}\t{host}{h_pad}\t{addr}{a_pad}\t[{tag}]{t_pad}\t{zone}{z_pad}\t{capacity}\tlast seen: {last_seen}s ago",
-					id=adv.id,
-					host=adv.state_info.hostname,
-					addr=adv.addr,
-					tag=cfg.tag,
-					zone=cfg.zone,
-					capacity=cfg.capacity_string(),
-					last_seen=(now_msec() - adv.last_seen) / 1000,
-					h_pad=" ".repeat(hostname_len - adv.state_info.hostname.len()),
-					a_pad=" ".repeat(addr_len - adv.addr.to_string().len()),
-					t_pad=" ".repeat(tag_len - cfg.tag.len()),
-					z_pad=" ".repeat(zone_len - cfg.zone.len()),
-				);
-			}
-		}
-		let (tag_len, zone_len) = config
-			.members
-			.iter()
-			.filter(|(&id, _)| !status.iter().any(|x| x.id == id))
-			.map(|(_, cfg)| (cfg.tag.len(), cfg.zone.len()))
-			.fold((0, 0), |(t, z), (mt, mz)| (max(t, mt), max(z, mz)));
-
-		for (id, cfg) in config.members.iter() {
-			if !status.iter().any(|x| x.id == *id) {
-				println!(
-					"{id:?}\t{tag}{t_pad}\t{zone}{z_pad}\t{capacity}\tnever seen",
-					id = id,
-					tag = cfg.tag,
-					zone = cfg.zone,
-					capacity = cfg.capacity_string(),
-					t_pad = " ".repeat(tag_len - cfg.tag.len()),
-					z_pad = " ".repeat(zone_len - cfg.zone.len()),
-				);
-			}
-		}
-	}
-
-	Ok(())
-}
-
-pub fn find_matching_node(
-	cand: impl std::iter::Iterator<Item = Uuid>,
-	pattern: &str,
-) -> Result<Uuid, Error> {
-	let mut candidates = vec![];
-	for c in cand {
-		if hex::encode(&c).starts_with(&pattern) {
-			candidates.push(c);
-		}
-	}
-	if candidates.len() != 1 {
-		Err(Error::Message(format!(
-			"{} nodes match '{}'",
-			candidates.len(),
-			pattern,
-		)))
-	} else {
-		Ok(candidates[0])
-	}
-}
-
-pub async fn cmd_configure(
-	rpc_cli: RpcAddrClient<Message>,
-	rpc_host: SocketAddr,
-	args: ConfigureNodeOpt,
-) -> Result<(), Error> {
-	let status = match rpc_cli
-		.call(&rpc_host, &Message::PullStatus, ADMIN_RPC_TIMEOUT)
-		.await??
-	{
-		Message::AdvertiseNodesUp(nodes) => nodes,
-		resp => return Err(Error::Message(format!("Invalid RPC response: {:?}", resp))),
-	};
-
-	let added_node = find_matching_node(status.iter().map(|x| x.id), &args.node_id)?;
-
-	let mut config = match rpc_cli
-		.call(&rpc_host, &Message::PullConfig, ADMIN_RPC_TIMEOUT)
-		.await??
-	{
-		Message::AdvertiseConfig(cfg) => cfg,
-		resp => return Err(Error::Message(format!("Invalid RPC response: {:?}", resp))),
-	};
-
-	for replaced in args.replace.iter() {
-		let replaced_node = find_matching_node(config.members.keys().cloned(), replaced)?;
-		if config.members.remove(&replaced_node).is_none() {
-			return Err(Error::Message(format!(
-				"Cannot replace node {:?} as it is not in current configuration",
-				replaced_node
-			)));
-		}
-	}
-
-	if args.capacity.is_some() && args.gateway {
-		return Err(Error::Message(
-				"-c and -g are mutually exclusive, please configure node either with c>0 to act as a storage node or with -g to act as a gateway node".into()));
-	}
-	if args.capacity == Some(0) {
-		return Err(Error::Message("Invalid capacity value: 0".into()));
-	}
-
-	let new_entry = match config.members.get(&added_node) {
-		None => {
-			let capacity = match args.capacity {
-				Some(c) => Some(c),
-				None if args.gateway => None,
-				_ => return Err(Error::Message(
-						"Please specify a capacity with the -c flag, or set node explicitly as gateway with -g".into())),
-			};
-			NetworkConfigEntry {
-				zone: args.zone.expect("Please specifiy a zone with the -z flag"),
-				capacity,
-				tag: args.tag.unwrap_or_default(),
-			}
-		}
-		Some(old) => {
-			let capacity = match args.capacity {
-				Some(c) => Some(c),
-				None if args.gateway => None,
-				_ => old.capacity,
-			};
-			NetworkConfigEntry {
-				zone: args.zone.unwrap_or_else(|| old.zone.to_string()),
-				capacity,
-				tag: args.tag.unwrap_or_else(|| old.tag.to_string()),
-			}
-		}
-	};
-
-	config.members.insert(added_node, new_entry);
-	config.version += 1;
-
-	rpc_cli
-		.call(
-			&rpc_host,
-			&Message::AdvertiseConfig(config),
-			ADMIN_RPC_TIMEOUT,
-		)
-		.await??;
-	Ok(())
-}
-
-pub async fn cmd_remove(
-	rpc_cli: RpcAddrClient<Message>,
-	rpc_host: SocketAddr,
-	args: RemoveNodeOpt,
-) -> Result<(), Error> {
-	let mut config = match rpc_cli
-		.call(&rpc_host, &Message::PullConfig, ADMIN_RPC_TIMEOUT)
-		.await??
-	{
-		Message::AdvertiseConfig(cfg) => cfg,
-		resp => return Err(Error::Message(format!("Invalid RPC response: {:?}", resp))),
-	};
-
-	let deleted_node = find_matching_node(config.members.keys().cloned(), &args.node_id)?;
-
-	if !args.yes {
-		return Err(Error::Message(format!(
-			"Add the flag --yes to really remove {:?} from the cluster",
-			deleted_node
-		)));
-	}
-
-	config.members.remove(&deleted_node);
-	config.version += 1;
-
-	rpc_cli
-		.call(
-			&rpc_host,
-			&Message::AdvertiseConfig(config),
-			ADMIN_RPC_TIMEOUT,
-		)
-		.await??;
-	Ok(())
-}
-
-pub async fn cmd_admin(
-	rpc_cli: RpcAddrClient<AdminRpc>,
-	rpc_host: SocketAddr,
-	args: AdminRpc,
-) -> Result<(), Error> {
-	match rpc_cli.call(&rpc_host, args, ADMIN_RPC_TIMEOUT).await?? {
-		AdminRpc::Ok(msg) => {
-			println!("{}", msg);
-		}
-		AdminRpc::BucketList(bl) => {
-			println!("List of buckets:");
-			for bucket in bl {
-				println!("{}", bucket);
-			}
-		}
-		AdminRpc::BucketInfo(bucket) => {
-			print_bucket_info(&bucket);
-		}
-		AdminRpc::KeyList(kl) => {
-			println!("List of keys:");
-			for key in kl {
-				println!("{}\t{}", key.0, key.1);
-			}
-		}
-		AdminRpc::KeyInfo(key) => {
-			print_key_info(&key);
-		}
-		r => {
-			error!("Unexpected response: {:?}", r);
-		}
-	}
-	Ok(())
-}
-
-fn print_key_info(key: &Key) {
-	println!("Key name: {}", key.name.get());
-	println!("Key ID: {}", key.key_id);
-	println!("Secret key: {}", key.secret_key);
-	if key.deleted.get() {
-		println!("Key is deleted.");
-	} else {
-		println!("Authorized buckets:");
-		for (b, _, perm) in key.authorized_buckets.items().iter() {
-			println!("- {} R:{} W:{}", b, perm.allow_read, perm.allow_write);
-		}
-	}
-}
-
-fn print_bucket_info(bucket: &Bucket) {
-	println!("Bucket name: {}", bucket.name);
-	match bucket.state.get() {
-		BucketState::Deleted => println!("Bucket is deleted."),
-		BucketState::Present(p) => {
-			println!("Authorized keys:");
-			for (k, _, perm) in p.authorized_keys.items().iter() {
-				println!("- {} R:{} W:{}", k, perm.allow_read, perm.allow_write);
-			}
-			println!("Website access: {}", p.website.get());
-		}
-	};
-}

src/garage/cli/cmd.rs

@@ -0,0 +1,184 @@
+use std::collections::HashSet;
+
+use garage_util::error::*;
+
+use garage_rpc::layout::*;
+use garage_rpc::system::*;
+use garage_rpc::*;
+
+use garage_model::helper::error::Error as HelperError;
+
+use crate::admin::*;
+use crate::cli::*;
+
+pub async fn cli_command_dispatch(
+	cmd: Command,
+	system_rpc_endpoint: &Endpoint<SystemRpc, ()>,
+	admin_rpc_endpoint: &Endpoint<AdminRpc, ()>,
+	rpc_host: NodeID,
+) -> Result<(), HelperError> {
+	match cmd {
+		Command::Status => Ok(cmd_status(system_rpc_endpoint, rpc_host).await?),
+		Command::Node(NodeOperation::Connect(connect_opt)) => {
+			Ok(cmd_connect(system_rpc_endpoint, rpc_host, connect_opt).await?)
+		}
+		Command::Layout(layout_opt) => {
+			Ok(cli_layout_command_dispatch(layout_opt, system_rpc_endpoint, rpc_host).await?)
+		}
+		Command::Bucket(bo) => {
+			cmd_admin(admin_rpc_endpoint, rpc_host, AdminRpc::BucketOperation(bo)).await
+		}
+		Command::Key(ko) => {
+			cmd_admin(admin_rpc_endpoint, rpc_host, AdminRpc::KeyOperation(ko)).await
+		}
+		Command::Migrate(mo) => {
+			cmd_admin(admin_rpc_endpoint, rpc_host, AdminRpc::Migrate(mo)).await
+		}
+		Command::Repair(ro) => {
+			cmd_admin(admin_rpc_endpoint, rpc_host, AdminRpc::LaunchRepair(ro)).await
+		}
+		Command::Stats(so) => cmd_admin(admin_rpc_endpoint, rpc_host, AdminRpc::Stats(so)).await,
+		_ => unreachable!(),
+	}
+}
+
+pub async fn cmd_status(rpc_cli: &Endpoint<SystemRpc, ()>, rpc_host: NodeID) -> Result<(), Error> {
+	let status = match rpc_cli
+		.call(&rpc_host, &SystemRpc::GetKnownNodes, PRIO_NORMAL)
+		.await??
+	{
+		SystemRpc::ReturnKnownNodes(nodes) => nodes,
+		resp => return Err(Error::Message(format!("Invalid RPC response: {:?}", resp))),
+	};
+	let layout = fetch_layout(rpc_cli, rpc_host).await?;
+
+	println!("==== HEALTHY NODES ====");
+	let mut healthy_nodes = vec!["ID\tHostname\tAddress\tTags\tZone\tCapacity".to_string()];
+	for adv in status.iter().filter(|adv| adv.is_up) {
+		match layout.roles.get(&adv.id) {
+			Some(NodeRoleV(Some(cfg))) => {
+				healthy_nodes.push(format!(
+					"{id:?}\t{host}\t{addr}\t[{tags}]\t{zone}\t{capacity}",
+					id = adv.id,
+					host = adv.status.hostname,
+					addr = adv.addr,
+					tags = cfg.tags.join(","),
+					zone = cfg.zone,
+					capacity = cfg.capacity_string(),
+				));
+			}
+			_ => {
+				let new_role = match layout.staging.get(&adv.id) {
+					Some(NodeRoleV(Some(_))) => "(pending)",
+					_ => "NO ROLE ASSIGNED",
+				};
+				healthy_nodes.push(format!(
+					"{id:?}\t{h}\t{addr}\t{new_role}",
+					id = adv.id,
+					h = adv.status.hostname,
+					addr = adv.addr,
+					new_role = new_role,
+				));
+			}
+		}
+	}
+	format_table(healthy_nodes);
+
+	let status_keys = status.iter().map(|adv| adv.id).collect::<HashSet<_>>();
+	let failure_case_1 = status.iter().any(|adv| !adv.is_up);
+	let failure_case_2 = layout
+		.roles
+		.items()
+		.iter()
+		.filter(|(_, _, v)| v.0.is_some())
+		.any(|(id, _, _)| !status_keys.contains(id));
+	if failure_case_1 || failure_case_2 {
+		println!("\n==== FAILED NODES ====");
+		let mut failed_nodes =
+			vec!["ID\tHostname\tAddress\tTags\tZone\tCapacity\tLast seen".to_string()];
+		for adv in status.iter().filter(|adv| !adv.is_up) {
+			if let Some(NodeRoleV(Some(cfg))) = layout.roles.get(&adv.id) {
+				failed_nodes.push(format!(
+					"{id:?}\t{host}\t{addr}\t[{tags}]\t{zone}\t{capacity}\t{last_seen}",
+					id = adv.id,
+					host = adv.status.hostname,
+					addr = adv.addr,
+					tags = cfg.tags.join(","),
+					zone = cfg.zone,
+					capacity = cfg.capacity_string(),
+					last_seen = adv
+						.last_seen_secs_ago
+						.map(|s| format!("{}s ago", s))
+						.unwrap_or_else(|| "never seen".into()),
+				));
+			}
+		}
+		for (id, _, role_v) in layout.roles.items().iter() {
+			if let NodeRoleV(Some(cfg)) = role_v {
+				if !status_keys.contains(id) {
+					failed_nodes.push(format!(
+						"{id:?}\t??\t??\t[{tags}]\t{zone}\t{capacity}\tnever seen",
+						id = id,
+						tags = cfg.tags.join(","),
+						zone = cfg.zone,
+						capacity = cfg.capacity_string(),
+					));
+				}
+			}
+		}
+		format_table(failed_nodes);
+	}
+
+	if print_staging_role_changes(&layout) {
+		println!();
+		println!("Please use `garage layout show` to check the proposed new layout and apply it.");
+		println!();
+	}
+
+	Ok(())
+}
+
+pub async fn cmd_connect(
+	rpc_cli: &Endpoint<SystemRpc, ()>,
+	rpc_host: NodeID,
+	args: ConnectNodeOpt,
+) -> Result<(), Error> {
+	match rpc_cli
+		.call(&rpc_host, &SystemRpc::Connect(args.node), PRIO_NORMAL)
+		.await??
+	{
+		SystemRpc::Ok => {
+			println!("Success.");
+			Ok(())
+		}
+		m => Err(Error::unexpected_rpc_message(m)),
+	}
+}
+
+pub async fn cmd_admin(
+	rpc_cli: &Endpoint<AdminRpc, ()>,
+	rpc_host: NodeID,
+	args: AdminRpc,
+) -> Result<(), HelperError> {
+	match rpc_cli.call(&rpc_host, &args, PRIO_NORMAL).await?? {
+		AdminRpc::Ok(msg) => {
+			println!("{}", msg);
+		}
+		AdminRpc::BucketList(bl) => {
+			print_bucket_list(bl);
+		}
+		AdminRpc::BucketInfo(bucket, rk) => {
+			print_bucket_info(&bucket, &rk);
+		}
+		AdminRpc::KeyList(kl) => {
+			print_key_list(kl);
+		}
+		AdminRpc::KeyInfo(key, rb) => {
+			print_key_info(&key, &rb);
+		}
+		r => {
+			error!("Unexpected response: {:?}", r);
+		}
+	}
+	Ok(())
+}

src/garage/cli/init.rs

@@ -0,0 +1,67 @@
+use std::path::PathBuf;
+
+use log::warn;
+
+use garage_util::error::*;
+
+pub const READ_KEY_ERROR: &str = "Unable to read node key. It will be generated by your garage node the first time is it launched. Ensure that your garage node is currently running. (The node key is supposed to be stored in your metadata directory.)";
+
+pub fn node_id_command(config_file: PathBuf, quiet: bool) -> Result<(), Error> {
+	let config = garage_util::config::read_config(config_file.clone()).err_context(format!(
+		"Unable to read configuration file {}",
+		config_file.to_string_lossy(),
+	))?;
+
+	let node_id =
+		garage_rpc::system::read_node_id(&config.metadata_dir).err_context(READ_KEY_ERROR)?;
+
+	let idstr = if let Some(addr) = config.rpc_public_addr {
+		let idstr = format!("{}@{}", hex::encode(&node_id), addr);
+		println!("{}", idstr);
+		idstr
+	} else {
+		let idstr = hex::encode(&node_id);
+		println!("{}", idstr);
+
+		if !quiet {
+			warn!("WARNING: I don't know the public address to reach this node.");
+			warn!("In all of the instructions below, replace 127.0.0.1:{} by the appropriate address and port.", config.rpc_bind_addr.port());
+		}
+
+		format!("{}@127.0.0.1:{}", idstr, config.rpc_bind_addr.port())
+	};
+
+	if !quiet {
+		eprintln!();
+		eprintln!(
+			"To instruct a node to connect to this node, run the following command on that node:"
+		);
+		eprintln!("    garage [-c <config file path>] node connect {}", idstr);
+		eprintln!();
+		eprintln!("Or instruct them to connect from here by running:");
+		eprintln!(
+			"    garage -c {} -h <remote node> node connect {}",
+			config_file.to_string_lossy(),
+			idstr
+		);
+		eprintln!(
+			"where <remote_node> is their own node identifier in the format: <pubkey>@<ip>:<port>"
+		);
+		eprintln!();
+		eprintln!("This node identifier can also be added as a bootstrap node in other node's garage.toml files:");
+		eprintln!("    bootstrap_peers = [");
+		eprintln!("        \"{}\",", idstr);
+		eprintln!("        ...");
+		eprintln!("    ]");
+		eprintln!();
+		eprintln!(
+			r#"Security notice: Garage's intra-cluster communications are secured primarily by the shared
+secret value rpc_secret.  However, an attacker that knows rpc_secret (for example if it
+leaks) cannot connect if they do not know any of the identifiers of the nodes in the
+cluster. It is thus a good security measure to try to keep them secret if possible.
+				  "#
+		);
+	}
+
+	Ok(())
+}

src/garage/cli/layout.rs

@@ -0,0 +1,340 @@
+use garage_util::crdt::Crdt;
+use garage_util::data::*;
+use garage_util::error::*;
+
+use garage_rpc::layout::*;
+use garage_rpc::system::*;
+use garage_rpc::*;
+
+use crate::cli::*;
+
+pub async fn cli_layout_command_dispatch(
+	cmd: LayoutOperation,
+	system_rpc_endpoint: &Endpoint<SystemRpc, ()>,
+	rpc_host: NodeID,
+) -> Result<(), Error> {
+	match cmd {
+		LayoutOperation::Assign(configure_opt) => {
+			cmd_assign_role(system_rpc_endpoint, rpc_host, configure_opt).await
+		}
+		LayoutOperation::Remove(remove_opt) => {
+			cmd_remove_role(system_rpc_endpoint, rpc_host, remove_opt).await
+		}
+		LayoutOperation::Show => cmd_show_layout(system_rpc_endpoint, rpc_host).await,
+		LayoutOperation::Apply(apply_opt) => {
+			cmd_apply_layout(system_rpc_endpoint, rpc_host, apply_opt).await
+		}
+		LayoutOperation::Revert(revert_opt) => {
+			cmd_revert_layout(system_rpc_endpoint, rpc_host, revert_opt).await
+		}
+	}
+}
+
+pub async fn cmd_assign_role(
+	rpc_cli: &Endpoint<SystemRpc, ()>,
+	rpc_host: NodeID,
+	args: AssignRoleOpt,
+) -> Result<(), Error> {
+	let status = match rpc_cli
+		.call(&rpc_host, &SystemRpc::GetKnownNodes, PRIO_NORMAL)
+		.await??
+	{
+		SystemRpc::ReturnKnownNodes(nodes) => nodes,
+		resp => return Err(Error::Message(format!("Invalid RPC response: {:?}", resp))),
+	};
+
+	let added_node = find_matching_node(status.iter().map(|adv| adv.id), &args.node_id)?;
+
+	let mut layout = fetch_layout(rpc_cli, rpc_host).await?;
+
+	let mut roles = layout.roles.clone();
+	roles.merge(&layout.staging);
+
+	for replaced in args.replace.iter() {
+		let replaced_node = find_matching_node(layout.node_ids().iter().cloned(), replaced)?;
+		match roles.get(&replaced_node) {
+			Some(NodeRoleV(Some(_))) => {
+				layout
+					.staging
+					.merge(&roles.update_mutator(replaced_node, NodeRoleV(None)));
+			}
+			_ => {
+				return Err(Error::Message(format!(
+					"Cannot replace node {:?} as it is not currently in planned layout",
+					replaced_node
+				)));
+			}
+		}
+	}
+
+	if args.capacity.is_some() && args.gateway {
+		return Err(Error::Message(
+				"-c and -g are mutually exclusive, please configure node either with c>0 to act as a storage node or with -g to act as a gateway node".into()));
+	}
+	if args.capacity == Some(0) {
+		return Err(Error::Message("Invalid capacity value: 0".into()));
+	}
+
+	let new_entry = match roles.get(&added_node) {
+		Some(NodeRoleV(Some(old))) => {
+			let capacity = match args.capacity {
+				Some(c) => Some(c),
+				None if args.gateway => None,
+				None => old.capacity,
+			};
+			let tags = if args.tags.is_empty() {
+				old.tags.clone()
+			} else {
+				args.tags
+			};
+			NodeRole {
+				zone: args.zone.unwrap_or_else(|| old.zone.to_string()),
+				capacity,
+				tags,
+			}
+		}
+		_ => {
+			let capacity = match args.capacity {
+				Some(c) => Some(c),
+				None if args.gateway => None,
+				None => return Err(Error::Message(
+						"Please specify a capacity with the -c flag, or set node explicitly as gateway with -g".into())),
+			};
+			NodeRole {
+				zone: args.zone.ok_or("Please specifiy a zone with the -z flag")?,
+				capacity,
+				tags: args.tags,
+			}
+		}
+	};
+
+	layout
+		.staging
+		.merge(&roles.update_mutator(added_node, NodeRoleV(Some(new_entry))));
+
+	send_layout(rpc_cli, rpc_host, layout).await?;
+
+	println!("Role change is staged but not yet commited.");
+	println!("Use `garage layout show` to view staged role changes,");
+	println!("and `garage layout apply` to enact staged changes.");
+	Ok(())
+}
+
+pub async fn cmd_remove_role(
+	rpc_cli: &Endpoint<SystemRpc, ()>,
+	rpc_host: NodeID,
+	args: RemoveRoleOpt,
+) -> Result<(), Error> {
+	let mut layout = fetch_layout(rpc_cli, rpc_host).await?;
+
+	let mut roles = layout.roles.clone();
+	roles.merge(&layout.staging);
+
+	let deleted_node =
+		find_matching_node(roles.items().iter().map(|(id, _, _)| *id), &args.node_id)?;
+
+	layout
+		.staging
+		.merge(&roles.update_mutator(deleted_node, NodeRoleV(None)));
+
+	send_layout(rpc_cli, rpc_host, layout).await?;
+
+	println!("Role removal is staged but not yet commited.");
+	println!("Use `garage layout show` to view staged role changes,");
+	println!("and `garage layout apply` to enact staged changes.");
+	Ok(())
+}
+
+pub async fn cmd_show_layout(
+	rpc_cli: &Endpoint<SystemRpc, ()>,
+	rpc_host: NodeID,
+) -> Result<(), Error> {
+	let mut layout = fetch_layout(rpc_cli, rpc_host).await?;
+
+	println!("==== CURRENT CLUSTER LAYOUT ====");
+	if !print_cluster_layout(&layout) {
+		println!("No nodes currently have a role in the cluster.");
+		println!("See `garage status` to view available nodes.");
+	}
+	println!();
+	println!("Current cluster layout version: {}", layout.version);
+
+	if print_staging_role_changes(&layout) {
+		layout.roles.merge(&layout.staging);
+
+		println!();
+		println!("==== NEW CLUSTER LAYOUT AFTER APPLYING CHANGES ====");
+		if !print_cluster_layout(&layout) {
+			println!("No nodes have a role in the new layout.");
+		}
+		println!();
+
+		// this will print the stats of what partitions
+		// will move around when we apply
+		if layout.calculate_partition_assignation() {
+			println!("To enact the staged role changes, type:");
+			println!();
+			println!("    garage layout apply --version {}", layout.version + 1);
+			println!();
+			println!(
+				"You can also revert all proposed changes with: garage layout revert --version {}",
+				layout.version + 1
+			);
+		} else {
+			println!("Not enough nodes have an assigned role to maintain enough copies of data.");
+			println!("This new layout cannot yet be applied.");
+		}
+	}
+
+	Ok(())
+}
+
+pub async fn cmd_apply_layout(
+	rpc_cli: &Endpoint<SystemRpc, ()>,
+	rpc_host: NodeID,
+	apply_opt: ApplyLayoutOpt,
+) -> Result<(), Error> {
+	let mut layout = fetch_layout(rpc_cli, rpc_host).await?;
+
+	match apply_opt.version {
+		None => {
+			println!("Please pass the --version flag to ensure that you are writing the correct version of the cluster layout.");
+			println!("To know the correct value of the --version flag, invoke `garage layout show` and review the proposed changes.");
+			return Err(Error::Message("--version flag is missing".into()));
+		}
+		Some(v) => {
+			if v != layout.version + 1 {
+				return Err(Error::Message("Invalid value of --version flag".into()));
+			}
+		}
+	}
+
+	layout.roles.merge(&layout.staging);
+
+	if !layout.calculate_partition_assignation() {
+		return Err(Error::Message("Could not calculate new assignation of partitions to nodes. This can happen if there are less nodes than the desired number of copies of your data (see the replication_mode configuration parameter).".into()));
+	}
+
+	layout.staging.clear();
+	layout.staging_hash = blake2sum(&rmp_to_vec_all_named(&layout.staging).unwrap()[..]);
+
+	layout.version += 1;
+
+	send_layout(rpc_cli, rpc_host, layout).await?;
+
+	println!("New cluster layout with updated role assignation has been applied in cluster.");
+	println!("Data will now be moved around between nodes accordingly.");
+
+	Ok(())
+}
+
+pub async fn cmd_revert_layout(
+	rpc_cli: &Endpoint<SystemRpc, ()>,
+	rpc_host: NodeID,
+	revert_opt: RevertLayoutOpt,
+) -> Result<(), Error> {
+	let mut layout = fetch_layout(rpc_cli, rpc_host).await?;
+
+	match revert_opt.version {
+		None => {
+			println!("Please pass the --version flag to ensure that you are writing the correct version of the cluster layout.");
+			println!("To know the correct value of the --version flag, invoke `garage layout show` and review the proposed changes.");
+			return Err(Error::Message("--version flag is missing".into()));
+		}
+		Some(v) => {
+			if v != layout.version + 1 {
+				return Err(Error::Message("Invalid value of --version flag".into()));
+			}
+		}
+	}
+
+	layout.staging.clear();
+	layout.staging_hash = blake2sum(&rmp_to_vec_all_named(&layout.staging).unwrap()[..]);
+
+	layout.version += 1;
+
+	send_layout(rpc_cli, rpc_host, layout).await?;
+
+	println!("All proposed role changes in cluster layout have been canceled.");
+	Ok(())
+}
+
+// --- utility ---
+
+pub async fn fetch_layout(
+	rpc_cli: &Endpoint<SystemRpc, ()>,
+	rpc_host: NodeID,
+) -> Result<ClusterLayout, Error> {
+	match rpc_cli
+		.call(&rpc_host, &SystemRpc::PullClusterLayout, PRIO_NORMAL)
+		.await??
+	{
+		SystemRpc::AdvertiseClusterLayout(t) => Ok(t),
+		resp => Err(Error::Message(format!("Invalid RPC response: {:?}", resp))),
+	}
+}
+
+pub async fn send_layout(
+	rpc_cli: &Endpoint<SystemRpc, ()>,
+	rpc_host: NodeID,
+	layout: ClusterLayout,
+) -> Result<(), Error> {
+	rpc_cli
+		.call(
+			&rpc_host,
+			&SystemRpc::AdvertiseClusterLayout(layout),
+			PRIO_NORMAL,
+		)
+		.await??;
+	Ok(())
+}
+
+pub fn print_cluster_layout(layout: &ClusterLayout) -> bool {
+	let mut table = vec!["ID\tTags\tZone\tCapacity".to_string()];
+	for (id, _, role) in layout.roles.items().iter() {
+		let role = match &role.0 {
+			Some(r) => r,
+			_ => continue,
+		};
+		let tags = role.tags.join(",");
+		table.push(format!(
+			"{:?}\t{}\t{}\t{}",
+			id,
+			tags,
+			role.zone,
+			role.capacity_string()
+		));
+	}
+	if table.len() == 1 {
+		false
+	} else {
+		format_table(table);
+		true
+	}
+}
+
+pub fn print_staging_role_changes(layout: &ClusterLayout) -> bool {
+	if !layout.staging.items().is_empty() {
+		println!();
+		println!("==== STAGED ROLE CHANGES ====");
+		let mut table = vec!["ID\tTags\tZone\tCapacity".to_string()];
+		for (id, _, role) in layout.staging.items().iter() {
+			if let Some(role) = &role.0 {
+				let tags = role.tags.join(",");
+				table.push(format!(
+					"{:?}\t{}\t{}\t{}",
+					id,
+					tags,
+					role.zone,
+					role.capacity_string()
+				));
+			} else {
+				table.push(format!("{:?}\tREMOVED", id));
+			}
+		}
+		format_table(table);
+		true
+	} else {
+		false
+	}
+}