Split format_table into separate crate and reduce k2v-client dependencies

Reviewed-on: https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/571

Cargo.toml

@@ -11,10 +11,22 @@ members = [
 	"src/web",
 	"src/garage",
 	"src/k2v-client",
+	"src/format-table",
 ]
 
 default-members = ["src/garage"]
 
+[workspace.dependencies]
+format_table = { version = "0.1.0", path = "src/format-table" }
+garage_api = { version = "0.8.2", path = "src/api" }
+garage_block = { version = "0.8.2", path = "src/block" }
+garage_db = { version = "0.8.2", path = "src/db", default-features = false }
+garage_model = { version = "0.8.2", path = "src/model", default-features = false }
+garage_rpc = { version = "0.8.2", path = "src/rpc" }
+garage_table = { version = "0.8.2", path = "src/table" }
+garage_util = { version = "0.8.2", path = "src/util" }
+garage_web = { version = "0.8.2", path = "src/web" }
+
 [profile.dev]
 lto = "off"
 

doc/book/connect/cli.md

@@ -202,56 +202,11 @@ See its usage output for other commands available.
 ## Cyberduck & duck {#cyberduck}
 
 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!
+Within Cyberduck, a
+[Garage connection profile](https://docs.cyberduck.io/protocols/s3/garage/) is
+available within the `Preferences -> Profiles` section. This can enabled and
+then connections to Garage may be configured.
 
 ### Instuctions for the CLI
 

doc/book/cookbook/real-world.md

@@ -349,5 +349,5 @@ and is covered in the [quick start guide](@/documentation/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 S3-compatible applicatiosn to interact with Garage
+Configuring S3-compatible applications to interact with Garage
 is covered in the [Integrations](@/documentation/connect/_index.md) section.

doc/book/quick-start/_index.md

@@ -290,13 +290,13 @@ sourcing the right file.*
 aws s3 ls
 
 # list objects of a bucket
-aws s3 ls s3://my_files
+aws s3 ls s3://nextcloud-bucket
 
 # copy from your filesystem to garage
-aws s3 cp /proc/cpuinfo s3://my_files/cpuinfo.txt
+aws s3 cp /proc/cpuinfo s3://nextcloud-bucket/cpuinfo.txt
 
 # copy from garage to your filesystem
-aws s3 cp s3/my_files/cpuinfo.txt /tmp/cpuinfo.txt
+aws s3 cp s3://nextcloud-bucket/cpuinfo.txt /tmp/cpuinfo.txt
 ```
 
 Note that you can use `awscli` for more advanced operations like

flake.lock

@@ -41,12 +41,15 @@
       }
     },
     "flake-utils": {
+      "inputs": {
+        "systems": "systems"
+      },
       "locked": {
-        "lastModified": 1659877975,
-        "narHash": "sha256-zllb8aq3YO3h8B/U0/J1WBgAL8EX5yWf5pMj3G0NAmc=",
+        "lastModified": 1681202837,
+        "narHash": "sha256-H+Rh19JDwRtpVPAWp64F+rlEtxUWBAQW28eAi3SRSzg=",
         "owner": "numtide",
         "repo": "flake-utils",
-        "rev": "c0e246b9b83f637f4681389ecabcb2681b4f3af0",
+        "rev": "cfacdce06f30d2b68473a46042957675eebb3401",
         "type": "github"
       },
       "original": {
@@ -56,12 +59,15 @@
       }
     },
     "flake-utils_2": {
+      "inputs": {
+        "systems": "systems_2"
+      },
       "locked": {
-        "lastModified": 1667395993,
-        "narHash": "sha256-nuEHfE/LcWyuSWnS8t12N1wc105Qtau+/OdUAjtQ0rA=",
+        "lastModified": 1681202837,
+        "narHash": "sha256-H+Rh19JDwRtpVPAWp64F+rlEtxUWBAQW28eAi3SRSzg=",
         "owner": "numtide",
         "repo": "flake-utils",
-        "rev": "5aed5285a952e0b949eb3ba02c12fa4fcfef535f",
+        "rev": "cfacdce06f30d2b68473a46042957675eebb3401",
         "type": "github"
       },
       "original": {
@@ -72,44 +78,61 @@
     },
     "nixpkgs": {
       "locked": {
-        "lastModified": 1665657542,
-        "narHash": "sha256-mojxNyzbvmp8NtVtxqiHGhRfjCALLfk9i/Uup68Y5q8=",
+        "lastModified": 1682109806,
+        "narHash": "sha256-d9g7RKNShMLboTWwukM+RObDWWpHKaqTYXB48clBWXI=",
         "owner": "NixOS",
         "repo": "nixpkgs",
-        "rev": "a3073c49bc0163fea6a121c276f526837672b555",
+        "rev": "2362848adf8def2866fabbffc50462e929d7fffb",
+        "type": "github"
+      },
+      "original": {
+        "owner": "NixOS",
+        "ref": "nixpkgs-unstable",
+        "repo": "nixpkgs",
+        "type": "github"
+      }
+    },
+    "nixpkgs_2": {
+      "locked": {
+        "lastModified": 1682423271,
+        "narHash": "sha256-WHhl1GiOij1ob4cTLL+yhqr+vFOUH8E5wAX8Ir8fvjE=",
+        "owner": "NixOS",
+        "repo": "nixpkgs",
+        "rev": "94517a501434a627c5d9e72ac6e7f26174b978d3",
         "type": "github"
       },
       "original": {
         "owner": "NixOS",
         "repo": "nixpkgs",
-        "rev": "a3073c49bc0163fea6a121c276f526837672b555",
+        "rev": "94517a501434a627c5d9e72ac6e7f26174b978d3",
         "type": "github"
       }
     },
     "root": {
       "inputs": {
         "cargo2nix": "cargo2nix",
-        "flake-utils": "flake-utils_2",
-        "nixpkgs": "nixpkgs"
-      }
-    },
-    "rust-overlay": {
-      "inputs": {
+        "flake-compat": [
+          "cargo2nix",
+          "flake-compat"
+        ],
         "flake-utils": [
           "cargo2nix",
           "flake-utils"
         ],
-        "nixpkgs": [
-          "cargo2nix",
-          "nixpkgs"
-        ]
+        "nixpkgs": "nixpkgs_2"
+      }
+    },
+    "rust-overlay": {
+      "inputs": {
+        "flake-utils": "flake-utils_2",
+        "nixpkgs": "nixpkgs"
       },
       "locked": {
-        "lastModified": 1664247556,
-        "narHash": "sha256-J4vazHU3609ekn7dr+3wfqPo5WGlZVAgV7jfux352L0=",
+        "lastModified": 1682389182,
+        "narHash": "sha256-8t2nmFnH+8V48+IJsf8AK51ebXNlVbOSVYOpiqJKvJE=",
         "owner": "oxalica",
         "repo": "rust-overlay",
-        "rev": "524db9c9ea7bc7743bb74cdd45b6d46ea3fcc2ab",
+        "rev": "74f1a64dd28faeeb85ef081f32cad2989850322c",
         "type": "github"
       },
       "original": {
@@ -117,6 +140,36 @@
         "repo": "rust-overlay",
         "type": "github"
       }
+    },
+    "systems": {
+      "locked": {
+        "lastModified": 1681028828,
+        "narHash": "sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768=",
+        "owner": "nix-systems",
+        "repo": "default",
+        "rev": "da67096a3b9bf56a91d16901293e51ba5b49a27e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-systems",
+        "repo": "default",
+        "type": "github"
+      }
+    },
+    "systems_2": {
+      "locked": {
+        "lastModified": 1681028828,
+        "narHash": "sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768=",
+        "owner": "nix-systems",
+        "repo": "default",
+        "rev": "da67096a3b9bf56a91d16901293e51ba5b49a27e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-systems",
+        "repo": "default",
+        "type": "github"
+      }
     }
   },
   "root": "root",

flake.nix

@@ -2,16 +2,30 @@
   description =
     "Garage, an S3-compatible distributed object store for self-hosted deployments";
 
+  # Nixpkgs unstable as of 2023-04-25, has rustc v1.68
   inputs.nixpkgs.url =
-    "github:NixOS/nixpkgs/a3073c49bc0163fea6a121c276f526837672b555";
+    "github:NixOS/nixpkgs/94517a501434a627c5d9e72ac6e7f26174b978d3";
+
   inputs.cargo2nix = {
     # As of 2022-10-18: two small patches over unstable branch, one for clippy and one to fix feature detection
     url = "github:Alexis211/cargo2nix/a7a61179b66054904ef6a195d8da736eaaa06c36";
+
+    # As of 2023-04-25:
+    # - my two patches were merged into unstable (one for clippy and one to "fix" feature detection)
+    # - rustc v1.66
+    # url = "github:cargo2nix/cargo2nix/8fb57a670f7993bfc24099c33eb9c5abb51f29a2";
+
+    # Rust overlay as of 2023-04-25
+    inputs.rust-overlay.url =
+      "github:oxalica/rust-overlay/74f1a64dd28faeeb85ef081f32cad2989850322c";
+
     inputs.nixpkgs.follows = "nixpkgs";
   };
-  inputs.flake-utils.url = "github:numtide/flake-utils";
 
-  outputs = { self, nixpkgs, cargo2nix, flake-utils }:
+  inputs.flake-utils.follows = "cargo2nix/flake-utils";
+  inputs.flake-compat.follows = "cargo2nix/flake-compat";
+
+  outputs = { self, nixpkgs, cargo2nix, flake-utils, ... }:
     let
       git_version = self.lastModifiedDate;
       compile = import ./nix/compile.nix;

nix/common.nix

@@ -1,19 +1,14 @@
-rec {
-  # * Fixed dependencies
-  pkgsSrc = fetchTarball {
-    # As of 2022-10-13
+let
+  lock = builtins.fromJSON (builtins.readFile ../flake.lock);
+  flakeCompatRev = lock.nodes.flake-compat.locked.rev;
+  flakeCompat = fetchTarball {
     url =
-      "https://github.com/NixOS/nixpkgs/archive/a3073c49bc0163fea6a121c276f526837672b555.zip";
-    sha256 = "1bz632psfbpmicyzjb8b4265y50shylccvfm6ry6mgnv5hvz324s";
+      "https://github.com/edolstra/flake-compat/archive/${flakeCompatRev}.tar.gz";
+    sha256 = lock.nodes.flake-compat.locked.narHash;
   };
-  cargo2nixSrc = fetchGit {
-    # As of 2022-10-18: two small patches over unstable branch, one for clippy and one to fix feature detection
-    url = "https://github.com/Alexis211/cargo2nix";
-    ref = "custom_unstable";
-    rev = "a7a61179b66054904ef6a195d8da736eaaa06c36";
-  };
-
-  # * Shared objects
-  cargo2nix = import cargo2nixSrc;
+  flake = ((import flakeCompat) { src = ../.; }).defaultNix;
+in rec {
+  pkgsSrc = flake.inputs.nixpkgs;
+  cargo2nix = flake.inputs.cargo2nix;
   cargo2nixOverlay = cargo2nix.overlays.default;
 }

nix/compile.nix

@@ -32,7 +32,7 @@ let
   */
   toolchainOptions = if target == null || target == "x86_64-unknown-linux-musl"
   || target == "aarch64-unknown-linux-musl" then {
-    rustVersion = "1.63.0";
+    rustVersion = "1.68.0";
     extraRustComponents = [ "clippy" ];
   } else {
     rustToolchain = pkgs.symlinkJoin {

nix/kaniko.nix

@@ -1,13 +1,13 @@
 pkgs:
 pkgs.buildGoModule rec {
   pname = "kaniko";
-  version = "1.6.0";
+  version = "1.9.2";
 
   src = pkgs.fetchFromGitHub {
     owner = "GoogleContainerTools";
     repo = "kaniko";
     rev = "v${version}";
-    sha256 = "1fnclr556avxay6pvgw5ya3xbxfnf2gv4njq2hr4fd6fcjyslq5h";
+    sha256 = "dXQ0/o1qISv+sjNVIpfF85bkbM9sGOGwqVbWZpMWfMY=";
   };
 
   vendorSha256 = null;

script/helm/garage/Chart.yaml

@@ -15,10 +15,10 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 0.3.0
+version: 0.4.1
 
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to
 # follow Semantic Versioning. They should reflect the version the application is using.
 # It is recommended to use it with quotes.
-appVersion: "v0.7.2.1"
+appVersion: "v0.8.2"

script/helm/garage/values.yaml

@@ -4,9 +4,28 @@
 
 # Garage configuration. These values go to garage.toml
 garage:
+  # Can be changed for better performance on certain systems
+  # https://garagehq.deuxfleurs.fr/documentation/reference-manual/configuration/#db-engine-since-v0-8-0
+  dbEngine: "sled"
+
+  # Defaults is 1MB
+  # An increase can result in better performance in certain scenarios
+  # https://garagehq.deuxfleurs.fr/documentation/reference-manual/configuration/#block-size
+  blockSize: "1048576"
+
+  # Tuning parameters for the sled DB engine
+  # https://garagehq.deuxfleurs.fr/documentation/reference-manual/configuration/#sled-cache-capacity
+  sledCacheCapacity: "134217728"
+  sledFlushEveryMs: "2000"
+
   # Default to 3 replicas, see the replication_mode section at
-  # https://garagehq.deuxfleurs.fr/documentation/reference-manual/configuration/
+  # https://garagehq.deuxfleurs.fr/documentation/reference-manual/configuration/#replication-mode
   replicationMode: "3"
+
+  # zstd compression level of stored blocks
+  # https://garagehq.deuxfleurs.fr/documentation/reference-manual/configuration/#compression-level
+  compressionLevel: "1"
+
   rpcBindAddr: "[::]:3901"
   # If not given, a random secret will be generated and stored in a Secret object
   rpcSecret: ""
@@ -27,8 +46,19 @@ garage:
     metadata_dir = "/mnt/meta"
     data_dir = "/mnt/data"
 
+    db_engine = "{{ .Values.garage.dbEngine }}"
+
+    block_size = {{ .Values.garage.blockSize }}
+
+    {{- if eq .Values.garage.dbEngine "sled"}}
+    sled_cache_capacity = {{ .Values.garage.sledCacheCapacity }}
+    sled_flush_every_ms = {{ .Values.garage.sledFlushEveryMs }}
+    {{- end }}
+
     replication_mode = "{{ .Values.garage.replicationMode }}"
 
+    compression_level = {{ .Values.garage.compressionLevel }}
+
     rpc_bind_addr = "{{ .Values.garage.rpcBindAddr }}"
     # rpc_secret will be populated by the init container from a k8s secret object
     rpc_secret = "__RPC_SECRET_REPLACE__"
@@ -49,7 +79,7 @@ garage:
     bind_addr = "[::]:3902"
     root_domain = "{{ .Values.garage.s3.web.rootDomain }}"
     index = "{{ .Values.garage.s3.web.index }}"
-    
+
     [admin]
     api_bind_addr = "[::]:3903"
     {{- if .Values.monitoring.tracing.sink }}
@@ -211,4 +241,4 @@ monitoring:
       scrapeTimeout: 10s
       relabelings: []
   tracing:
-    sink: ""
+    sink: ""

src/api/Cargo.toml

@@ -14,11 +14,11 @@ path = "lib.rs"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-garage_model = { version = "0.8.2", path = "../model" }
-garage_table = { version = "0.8.2", path = "../table" }
-garage_block = { version = "0.8.2", path = "../block" }
-garage_util = { version = "0.8.2", path = "../util" }
-garage_rpc = { version = "0.8.2", path = "../rpc" }
+garage_model.workspace = true
+garage_table.workspace = true
+garage_block.workspace = true
+garage_util.workspace = true
+garage_rpc.workspace = true
 
 async-trait = "0.1.7"
 base64 = "0.21"
@@ -51,7 +51,7 @@ roxmltree = "0.14"
 serde = { version = "1.0", features = ["derive"] }
 serde_bytes = "0.11"
 serde_json = "1.0"
-quick-xml = { version = "0.21", features = [ "serialize" ] }
+quick-xml = { version = "0.26", features = [ "serialize" ] }
 url = "2.3"
 
 opentelemetry = "0.17"

src/api/admin/api_server.rs

@@ -105,7 +105,7 @@ impl AdminApiServer {
 		let bucket_id = self
 			.garage
 			.bucket_helper()
-			.resolve_global_bucket_name(&domain)
+			.resolve_global_bucket_name(domain)
 			.await?
 			.ok_or(HelperError::NoSuchBucket(domain.to_string()))?;
 

src/api/admin/bucket.rs

@@ -183,8 +183,8 @@ async fn bucket_info_results(
 				}
 			}),
 			keys: relevant_keys
-				.into_iter()
-				.map(|(_, key)| {
+				.into_values()
+				.map(|key| {
 					let p = key.state.as_option().unwrap();
 					GetBucketInfoKey {
 						access_key_id: key.key_id,

src/api/admin/key.rs

@@ -183,8 +183,8 @@ async fn key_info_results(garage: &Arc<Garage>, key: Key) -> Result<Response<Bod
 			create_bucket: *key_state.allow_create_bucket.get(),
 		},
 		buckets: relevant_buckets
-			.into_iter()
-			.map(|(_, bucket)| {
+			.into_values()
+			.map(|bucket| {
 				let state = bucket.state.as_option().unwrap();
 				KeyInfoBucketResult {
 					id: hex::encode(bucket.id),

src/api/generic_server.rs

@@ -128,7 +128,7 @@ impl<A: ApiHandler> ApiServer<A> {
 		let uri = req.uri().clone();
 
 		if let Ok(forwarded_for_ip_addr) =
-			forwarded_headers::handle_forwarded_for_headers(&req.headers())
+			forwarded_headers::handle_forwarded_for_headers(req.headers())
 		{
 			info!(
 				"{} (via {}) {} {}",

src/api/k2v/batch.rs

@@ -282,8 +282,8 @@ pub(crate) async fn handle_poll_range(
 	if let Some((items, seen_marker)) = resp {
 		let resp = PollRangeResponse {
 			items: items
-				.into_iter()
-				.map(|(_k, i)| ReadBatchResponseItem::from(i))
+				.into_values()
+				.map(ReadBatchResponseItem::from)
 				.collect::<Vec<_>>(),
 			seen_marker,
 		};

src/api/s3/get.rs

@@ -443,7 +443,7 @@ fn body_from_blocks_range(
 	// block.part_number, which is not the same in the case of a multipart upload)
 	let mut blocks: Vec<(VersionBlock, u64)> = Vec::with_capacity(std::cmp::min(
 		all_blocks.len(),
-		4 + ((end - begin) / std::cmp::max(all_blocks[0].1.size as u64, 1024)) as usize,
+		4 + ((end - begin) / std::cmp::max(all_blocks[0].1.size, 1024)) as usize,
 	));
 	let mut block_offset: u64 = 0;
 	for (_, b) in all_blocks.iter() {
@@ -454,7 +454,7 @@ fn body_from_blocks_range(
 		if block_offset < end && block_offset + b.size > begin {
 			blocks.push((*b, block_offset));
 		}
-		block_offset += b.size as u64;
+		block_offset += b.size;
 	}
 
 	let order_stream = OrderTag::stream();

src/block/Cargo.toml

@@ -14,10 +14,10 @@ path = "lib.rs"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-garage_db = { version = "0.8.2", path = "../db" }
-garage_rpc = { version = "0.8.2", path = "../rpc" }
-garage_util = { version = "0.8.2", path = "../util" }
-garage_table = { version = "0.8.2", path = "../table" }
+garage_db.workspace = true
+garage_rpc.workspace = true
+garage_util.workspace = true
+garage_table.workspace = true
 
 opentelemetry = "0.17"
 
@@ -37,7 +37,7 @@ serde_bytes = "0.11"
 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"] }
-tokio-util = { version = "0.6", features = ["io"] }
+tokio-util = { version = "0.7", features = ["io"] }
 
 [features]
 system-libs = [ "zstd/pkg-config" ]

src/block/manager.rs

@@ -600,12 +600,32 @@ impl BlockManager {
 	/// Utility: check if block is stored compressed. Error if block is not stored
 	async fn is_block_compressed(&self, hash: &Hash) -> Result<bool, Error> {
 		let mut path = self.block_path(hash);
-		path.set_extension("zst");
-		if fs::metadata(&path).await.is_ok() {
-			return Ok(true);
+
+		// If compression is disabled on node - check for the raw block
+		// first and then a compressed one (as compression may have been
+		// previously enabled).
+		match self.compression_level {
+			None => {
+				if fs::metadata(&path).await.is_ok() {
+					return Ok(false);
+				}
+
+				path.set_extension("zst");
+
+				fs::metadata(&path).await.map(|_| true).map_err(Into::into)
+			}
+			_ => {
+				path.set_extension("zst");
+
+				if fs::metadata(&path).await.is_ok() {
+					return Ok(true);
+				}
+
+				path.set_extension("");
+
+				fs::metadata(&path).await.map(|_| false).map_err(Into::into)
+			}
 		}
-		path.set_extension("");
-		fs::metadata(&path).await.map(|_| false).map_err(Into::into)
 	}
 
 	async fn lock_mutate(&self, hash: &Hash) -> MutexGuard<'_, BlockManagerLocked> {

src/block/repair.rs

@@ -220,14 +220,12 @@ fn randomize_next_scrub_run_time(timestamp: u64) -> u64 {
 	// Take SCRUB_INTERVAL and mix in a random interval of 10 days to attempt to
 	// balance scrub load across different cluster nodes.
 
-	let next_run_timestamp = timestamp
+	timestamp
 		+ SCRUB_INTERVAL
 			.saturating_add(Duration::from_secs(
 				rand::thread_rng().gen_range(0..3600 * 24 * 10),
 			))
-			.as_millis() as u64;
-
-	next_run_timestamp
+			.as_millis() as u64
 }
 
 impl Default for ScrubWorkerPersisted {
@@ -241,18 +239,14 @@ impl Default for ScrubWorkerPersisted {
 	}
 }
 
+#[derive(Default)]
 enum ScrubWorkerState {
 	Running(BlockStoreIterator),
 	Paused(BlockStoreIterator, u64), // u64 = time when to resume scrub
+	#[default]
 	Finished,
 }
 
-impl Default for ScrubWorkerState {
-	fn default() -> Self {
-		ScrubWorkerState::Finished
-	}
-}
-
 #[derive(Debug)]
 pub enum ScrubWorkerCommand {
 	Start,
@@ -281,6 +275,7 @@ impl ScrubWorker {
 			ScrubWorkerCommand::Start => {
 				self.work = match std::mem::take(&mut self.work) {
 					ScrubWorkerState::Finished => {
+						info!("Scrub worker initializing, now performing datastore scrub");
 						let iterator = BlockStoreIterator::new(&self.manager);
 						ScrubWorkerState::Running(iterator)
 					}
@@ -394,12 +389,21 @@ impl Worker for ScrubWorker {
 						.tranquilizer
 						.tranquilize_worker(self.persister.get_with(|p| p.tranquility)))
 				} else {
+					let now = now_msec();
+					let next_scrub_timestamp = randomize_next_scrub_run_time(now);
+
 					self.persister.set_with(|p| {
-						p.time_last_complete_scrub = now_msec();
-						p.time_next_run_scrub = randomize_next_scrub_run_time(now_msec());
+						p.time_last_complete_scrub = now;
+						p.time_next_run_scrub = next_scrub_timestamp;
 					})?;
 					self.work = ScrubWorkerState::Finished;
 					self.tranquilizer.clear();
+
+					info!(
+						"Datastore scrub completed, next scrub scheduled for {}",
+						msec_to_rfc3339(next_scrub_timestamp)
+					);
+
 					Ok(WorkerState::Idle)
 				}
 			}

src/db/lmdb_adapter.rs

@@ -349,6 +349,6 @@ pub fn recommended_map_size() -> usize {
 
 #[cfg(target_pointer_width = "32")]
 pub fn recommended_map_size() -> usize {
-	warn!("LMDB is not recommended on 32-bit systems, database size will be limited");
+	tracing::warn!("LMDB is not recommended on 32-bit systems, database size will be limited");
 	1usize << 30
 }

src/format-table/Cargo.toml

@@ -0,0 +1,12 @@
+[package]
+name = "format_table"
+version = "0.1.0"
+authors = ["Alex Auvolat <alex@adnab.me>"]
+edition = "2018"
+license = "AGPL-3.0"
+description = "Format tables with a stupid API"
+repository = "https://git.deuxfleurs.fr/Deuxfleurs/garage"
+readme = "README.md"
+
+[lib]
+path = "lib.rs"

src/format-table/README.md

@@ -0,0 +1,13 @@
+# `format_table`
+
+Format tables with a stupid API. [Documentation](https://docs.rs/format_table).
+
+Example:
+
+```rust
+let mut table = vec!["product\tquantity\tprice".to_string()];
+for (p, q, r) in [("tomato", 12, 15), ("potato", 10, 20), ("rice", 5, 12)] {
+	table.push(format!("{}\t{}\t{}", p, q, r));
+}
+format_table::format_table(table);
+```

src/format-table/lib.rs

@@ -1,3 +1,19 @@
+//! Format tables with a stupid API.
+//!
+//! Example:
+//!
+//! ```rust
+//! let mut table = vec!["product\tquantity\tprice".to_string()];
+//! for (p, q, r) in [("tomato", 12, 15), ("potato", 10, 20), ("rice", 5, 12)] {
+//! 	table.push(format!("{}\t{}\t{}", p, q, r));
+//! }
+//! format_table::format_table(table);
+//! ```
+//!
+//! A table to be formatted is a `Vec<String>`, containing one string per line.
+//! Table columns in each line are separated by a `\t` character.
+
+/// Format a table and return the result as a string.
 pub fn format_table_to_string(data: Vec<String>) -> String {
 	let data = data
 		.iter()
@@ -27,6 +43,7 @@ pub fn format_table_to_string(data: Vec<String>) -> String {
 	out
 }
 
+/// Format a table and print the result to stdout.
 pub fn format_table(data: Vec<String>) {
 	print!("{}", format_table_to_string(data));
 }

src/garage/Cargo.toml

@@ -21,19 +21,20 @@ path = "tests/lib.rs"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-garage_db = { version = "0.8.2", path = "../db" }
-garage_api = { version = "0.8.2", path = "../api" }
-garage_block = { version = "0.8.2", path = "../block" }
-garage_model = { version = "0.8.2", path = "../model" }
-garage_rpc = { version = "0.8.2", path = "../rpc" }
-garage_table = { version = "0.8.2", path = "../table" }
-garage_util = { version = "0.8.2", path = "../util" }
-garage_web = { version = "0.8.2", path = "../web" }
+format_table.workspace = true
+garage_db.workspace = true
+garage_api.workspace = true
+garage_block.workspace = true
+garage_model.workspace = true
+garage_rpc.workspace = true
+garage_table.workspace = true
+garage_util.workspace = true
+garage_web.workspace = true
 
 backtrace = "0.3"
 bytes = "1.0"
 bytesize = "1.1"
-timeago = "0.4"
+timeago = { version = "0.4", default-features = false }
 parse_duration = "2.1"
 hex = "0.4"
 tracing = { version = "0.1" }
@@ -41,6 +42,7 @@ tracing-subscriber = { version = "0.3", features = ["env-filter"] }
 rand = "0.8"
 async-trait = "0.1.7"
 sodiumoxide = { version = "0.2.5-0", package = "kuska-sodiumoxide" }
+git-version = "0.3.4"
 
 serde = { version = "1.0", default-features = false, features = ["derive", "rc"] }
 serde_bytes = "0.11"

src/garage/admin.rs

@@ -5,11 +5,11 @@ use std::sync::Arc;
 use async_trait::async_trait;
 use serde::{Deserialize, Serialize};
 
+use format_table::format_table_to_string;
 use garage_util::background::BackgroundRunner;
 use garage_util::crdt::*;
 use garage_util::data::*;
 use garage_util::error::Error as GarageError;
-use garage_util::formater::format_table_to_string;
 use garage_util::time::*;
 
 use garage_table::replication::*;

src/garage/cli/cmd.rs

@@ -1,8 +1,8 @@
 use std::collections::HashSet;
 use std::time::Duration;
 
+use format_table::format_table;
 use garage_util::error::*;
-use garage_util::formater::format_table;
 
 use garage_rpc::layout::*;
 use garage_rpc::system::*;

src/garage/cli/layout.rs

@@ -1,6 +1,6 @@
+use format_table::format_table;
 use garage_util::crdt::Crdt;
 use garage_util::error::*;
-use garage_util::formater::format_table;
 
 use garage_rpc::layout::*;
 use garage_rpc::system::*;

src/garage/cli/util.rs

@@ -1,11 +1,11 @@
 use std::collections::HashMap;
 use std::time::Duration;
 
+use format_table::format_table;
 use garage_util::background::*;
 use garage_util::crdt::*;
 use garage_util::data::*;
 use garage_util::error::*;
-use garage_util::formater::format_table;
 use garage_util::time::*;
 
 use garage_block::manager::BlockResyncErrorInfo;

src/garage/main.rs

@@ -108,6 +108,12 @@ async fn main() {
 	][..];
 	if let Some(git_version) = option_env!("GIT_VERSION") {
 		garage_util::version::init_version(git_version);
+	} else {
+		garage_util::version::init_version(git_version::git_version!(
+			prefix = "git:",
+			cargo_prefix = "cargo:",
+			fallback = "unknown"
+		));
 	}
 	garage_util::version::init_features(features);
 

src/k2v-client/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "k2v-client"
-version = "0.1.1"
+version = "0.0.2"
 authors = ["Trinity Pointard <trinity.pointard@gmail.com>", "Alex Auvolat <alex@adnab.me>"]
 edition = "2018"
 license = "AGPL-3.0"
@@ -19,15 +19,15 @@ hyper-rustls = { version = "0.23", default-features = false, features = [ "http1
 serde = "1.0"
 serde_json = "1.0"
 thiserror = "1.0"
-tokio = "1.24"
+tokio = { version = "1.0", default-features = false, features = ["rt", "rt-multi-thread", "io-util", "net", "time", "macros", "sync", "signal", "fs"] }
 
 # cli deps
 clap = { version = "4.1", optional = true, features = ["derive", "env"] }
-garage_util = { version = "0.8.2", path = "../util", optional = true }
+format_table = { workspace = true, optional = true }
 
 
 [features]
-cli = ["clap", "tokio/fs", "tokio/io-std", "garage_util"]
+cli = ["clap", "tokio/fs", "tokio/io-std", "format_table"]
 
 [lib]
 path = "lib.rs"

src/k2v-client/bin/k2v-cli.rs

@@ -2,9 +2,11 @@ use std::collections::BTreeMap;
 use std::process::exit;
 use std::time::Duration;
 
+use base64::prelude::*;
+
 use k2v_client::*;
 
-use garage_util::formater::format_table;
+use format_table::format_table;
 
 use rusoto_core::credential::AwsCredentials;
 use rusoto_core::Region;
@@ -155,7 +157,9 @@ impl Value {
 		if let Some(ref text) = self.text {
 			Ok(text.as_bytes().to_vec())
 		} else if let Some(ref b64) = self.b64 {
-			base64::decode(b64).map_err(|_| Error::Message("invalid base64 input".into()))
+			BASE64_STANDARD
+				.decode(b64)
+				.map_err(|_| Error::Message("invalid base64 input".into()))
 		} else if let Some(ref path) = self.file {
 			use tokio::io::AsyncReadExt;
 			if path == "-" {
@@ -230,7 +234,7 @@ impl ReadOutputKind {
 			for val in val.value {
 				match val {
 					K2vValue::Value(v) => {
-						println!("{}", base64::encode(&v))
+						println!("{}", BASE64_STANDARD.encode(&v))
 					}
 					K2vValue::Tombstone => {
 						println!();
@@ -249,7 +253,7 @@ impl ReadOutputKind {
 					if let Ok(string) = std::str::from_utf8(&v) {
 						println!("  utf-8: {}", string);
 					} else {
-						println!("  base64: {}", base64::encode(&v));
+						println!("  base64: {}", BASE64_STANDARD.encode(&v));
 					}
 				}
 				K2vValue::Tombstone => {
@@ -275,7 +279,7 @@ struct BatchOutputKind {
 impl BatchOutputKind {
 	fn display_human_output(&self, values: BTreeMap<String, CausalValue>) -> ! {
 		for (key, values) in values {
-			println!("key: {}", key);
+			println!("sort_key: {}", key);
 			let causality: String = values.causality.into();
 			println!("causality: {}", causality);
 			for value in values.value {
@@ -284,7 +288,7 @@ impl BatchOutputKind {
 						if let Ok(string) = std::str::from_utf8(&v) {
 							println!("  value(utf-8): {}", string);
 						} else {
-							println!("  value(base64): {}", base64::encode(&v));
+							println!("  value(base64): {}", BASE64_STANDARD.encode(&v));
 						}
 					}
 					K2vValue::Tombstone => {
@@ -520,7 +524,7 @@ async fn main() -> Result<(), Error> {
 						value
 							.as_object_mut()
 							.unwrap()
-							.insert("sort_key".to_owned(), k.into());
+							.insert("partition_key".to_owned(), k.into());
 						value
 					})
 					.collect::<Vec<_>>();
@@ -537,7 +541,7 @@ async fn main() -> Result<(), Error> {
 				}
 
 				let mut to_print = Vec::new();
-				to_print.push(format!("key:\tentries\tconflicts\tvalues\tbytes"));
+				to_print.push(format!("partition_key\tentries\tconflicts\tvalues\tbytes"));
 				for (k, v) in res.items {
 					to_print.push(format!(
 						"{}\t{}\t{}\t{}\t{}",

src/k2v-client/lib.rs

@@ -1,6 +1,7 @@
 use std::collections::BTreeMap;
 use std::time::Duration;
 
+use base64::prelude::*;
 use http::header::{ACCEPT, CONTENT_LENGTH, CONTENT_TYPE};
 use http::status::StatusCode;
 use http::HeaderMap;
@@ -375,7 +376,7 @@ impl K2vClient {
 					.unwrap_or_default();
 				let err_body_str = std::str::from_utf8(&err_body)
 					.map(String::from)
-					.unwrap_or_else(|_| base64::encode(&err_body));
+					.unwrap_or_else(|_| BASE64_STANDARD.encode(&err_body));
 
 				if s.is_client_error() || s.is_server_error() {
 					error!("Error response {}: {}", res.status, err_body_str);
@@ -408,7 +409,7 @@ impl K2vClient {
 			"Response body: {}",
 			std::str::from_utf8(&body)
 				.map(String::from)
-				.unwrap_or_else(|_| base64::encode(&body))
+				.unwrap_or_else(|_| BASE64_STANDARD.encode(&body))
 		);
 
 		Ok(Response {
@@ -482,9 +483,11 @@ impl<'de> Deserialize<'de> for K2vValue {
 	{
 		let val: Option<&str> = Option::deserialize(d)?;
 		Ok(match val {
-			Some(s) => {
-				K2vValue::Value(base64::decode(s).map_err(|_| DeError::custom("invalid base64"))?)
-			}
+			Some(s) => K2vValue::Value(
+				BASE64_STANDARD
+					.decode(s)
+					.map_err(|_| DeError::custom("invalid base64"))?,
+			),
 			None => K2vValue::Tombstone,
 		})
 	}
@@ -498,7 +501,7 @@ impl Serialize for K2vValue {
 		match self {
 			K2vValue::Tombstone => serializer.serialize_none(),
 			K2vValue::Value(v) => {
-				let b64 = base64::encode(v);
+				let b64 = BASE64_STANDARD.encode(v);
 				serializer.serialize_str(&b64)
 			}
 		}

src/model/Cargo.toml

@@ -14,11 +14,11 @@ path = "lib.rs"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-garage_db = { version = "0.8.2", default-features = false, path = "../db" }
-garage_rpc = { version = "0.8.2", path = "../rpc" }
-garage_table = { version = "0.8.2", path = "../table" }
-garage_block = { version = "0.8.2", path = "../block" }
-garage_util = { version = "0.8.2", path = "../util" }
+garage_db.workspace = true
+garage_rpc.workspace = true
+garage_table.workspace = true
+garage_block.workspace = true
+garage_util.workspace = true
 
 async-trait = "0.1.7"
 arc-swap = "1.0"

src/model/k2v/seen.rs

@@ -79,7 +79,7 @@ impl RangeSeenMarker {
 
 		let bytes = nonversioned_encode(&self)?;
 		let bytes = zstd::stream::encode_all(&mut &bytes[..], zstd::DEFAULT_COMPRESSION_LEVEL)?;
-		Ok(BASE64_STANDARD.encode(&bytes))
+		Ok(BASE64_STANDARD.encode(bytes))
 	}
 
 	/// Decode from msgpack+zstd+b64 representation, returns None on error.

src/rpc/Cargo.toml

@@ -14,7 +14,7 @@ path = "lib.rs"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-garage_util = { version = "0.8.2", path = "../util" }
+garage_util.workspace = true
 
 arc-swap = "1.0"
 bytes = "1.0"

src/table/Cargo.toml

@@ -14,9 +14,9 @@ path = "lib.rs"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-garage_db = { version = "0.8.2", path = "../db" }
-garage_rpc = { version = "0.8.2", path = "../rpc" }
-garage_util = { version = "0.8.2", path = "../util" }
+garage_db.workspace = true
+garage_rpc.workspace = true
+garage_util.workspace = true
 
 opentelemetry = "0.17"
 

src/table/data.rs

@@ -44,22 +44,22 @@ pub struct TableData<F: TableSchema, R: TableReplication> {
 impl<F: TableSchema, R: TableReplication> TableData<F, R> {
 	pub fn new(system: Arc<System>, instance: F, replication: R, db: &db::Db) -> Arc<Self> {
 		let store = db
-			.open_tree(&format!("{}:table", F::TABLE_NAME))
+			.open_tree(format!("{}:table", F::TABLE_NAME))
 			.expect("Unable to open DB tree");
 
 		let merkle_tree = db
-			.open_tree(&format!("{}:merkle_tree", F::TABLE_NAME))
+			.open_tree(format!("{}:merkle_tree", F::TABLE_NAME))
 			.expect("Unable to open DB Merkle tree tree");
 		let merkle_todo = db
-			.open_tree(&format!("{}:merkle_todo", F::TABLE_NAME))
+			.open_tree(format!("{}:merkle_todo", F::TABLE_NAME))
 			.expect("Unable to open DB Merkle TODO tree");
 
 		let insert_queue = db
-			.open_tree(&format!("{}:insert_queue", F::TABLE_NAME))
+			.open_tree(format!("{}:insert_queue", F::TABLE_NAME))
 			.expect("Unable to open insert queue DB tree");
 
 		let gc_todo = db
-			.open_tree(&format!("{}:gc_todo_v2", F::TABLE_NAME))
+			.open_tree(format!("{}:gc_todo_v2", F::TABLE_NAME))
 			.expect("Unable to open GC DB tree");
 		let gc_todo = CountedTree::new(gc_todo).expect("Cannot count gc_todo_v2");
 
@@ -90,7 +90,7 @@ impl<F: TableSchema, R: TableReplication> TableData<F, R> {
 
 	pub fn read_entry(&self, p: &F::P, s: &F::S) -> Result<Option<ByteBuf>, Error> {
 		let tree_key = self.tree_key(p, s);
-		if let Some(bytes) = self.store.get(&tree_key)? {
+		if let Some(bytes) = self.store.get(tree_key)? {
 			Ok(Some(ByteBuf::from(bytes.to_vec())))
 		} else {
 			Ok(None)
@@ -132,10 +132,10 @@ impl<F: TableSchema, R: TableReplication> TableData<F, R> {
 		}
 	}
 
-	fn read_range_aux<'a>(
+	fn read_range_aux(
 		&self,
 		partition_hash: Hash,
-		range: db::ValueIter<'a>,
+		range: db::ValueIter,
 		filter: &Option<F::Filter>,
 		limit: usize,
 	) -> Result<Vec<Arc<ByteBuf>>, Error> {

src/table/util.rs

@@ -34,8 +34,9 @@ impl DeletedFilter {
 	}
 }
 
-#[derive(Clone, Copy, Debug, Serialize, Deserialize, PartialEq, Eq)]
+#[derive(Clone, Copy, Debug, Default, Serialize, Deserialize, PartialEq, Eq)]
 pub enum EnumerationOrder {
+	#[default]
 	Forward,
 	Reverse,
 }
@@ -49,9 +50,3 @@ impl EnumerationOrder {
 		}
 	}
 }
-
-impl Default for EnumerationOrder {
-	fn default() -> Self {
-		EnumerationOrder::Forward
-	}
-}

src/util/Cargo.toml

@@ -14,7 +14,7 @@ path = "lib.rs"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-garage_db = { version = "0.8.2", path = "../db" }
+garage_db.workspace = true
 
 arc-swap = "1.0"
 async-trait = "0.1"
@@ -22,7 +22,6 @@ blake2 = "0.10"
 bytes = "1.0"
 digest = "0.10"
 err-derive = "0.3"
-git-version = "0.3.4"
 hexdump = "0.1"
 xxhash-rust = { version = "0.8", default-features = false, features = ["xxh3"] }
 hex = "0.4"

src/util/config.rs

@@ -223,7 +223,7 @@ fn secret_from_file(
 			#[cfg(unix)]
 			if std::env::var("GARAGE_ALLOW_WORLD_READABLE_SECRETS").as_deref() != Ok("true") {
 				use std::os::unix::fs::MetadataExt;
-				let metadata = std::fs::metadata(&file_path)?;
+				let metadata = std::fs::metadata(file_path)?;
 				if metadata.mode() & 0o077 != 0 {
 					return Err(format!("File {} is world-readable! (mode: 0{:o}, expected 0600)\nRefusing to start until this is fixed, or environment variable GARAGE_ALLOW_WORLD_READABLE_SECRETS is set to true.", file_path, metadata.mode()).into());
 				}

src/util/forwarded_headers.rs

@@ -13,7 +13,7 @@ pub fn handle_forwarded_for_headers(headers: &HeaderMap<HeaderValue>) -> Result<
 		.to_str()
 		.ok_or_message("Error parsing X-Forwarded-For header")?;
 
-	let client_ip = IpAddr::from_str(&forwarded_for_ip_str)
+	let client_ip = IpAddr::from_str(forwarded_for_ip_str)
 		.ok_or_message("Valid IP address not found in X-Forwarded-For header")?;
 
 	Ok(client_ip.to_string())

src/util/lib.rs

@@ -10,7 +10,6 @@ pub mod crdt;
 pub mod data;
 pub mod encode;
 pub mod error;
-pub mod formater;
 pub mod forwarded_headers;
 pub mod metrics;
 pub mod migrate;

src/util/version.rs

@@ -1,18 +1,14 @@
 use std::sync::Arc;
 
-use arc_swap::{ArcSwap, ArcSwapOption};
+use arc_swap::ArcSwapOption;
 
 lazy_static::lazy_static! {
-	static ref VERSION: ArcSwap<&'static str> = ArcSwap::new(Arc::new(git_version::git_version!(
-		prefix = "git:",
-		cargo_prefix = "cargo:",
-		fallback = "unknown"
-	)));
+	static ref VERSION: ArcSwapOption<&'static str> = ArcSwapOption::new(None);
 	static ref FEATURES: ArcSwapOption<&'static [&'static str]> = ArcSwapOption::new(None);
 }
 
 pub fn garage_version() -> &'static str {
-	&VERSION.load()
+	VERSION.load().as_ref().unwrap()
 }
 
 pub fn garage_features() -> Option<&'static [&'static str]> {
@@ -20,7 +16,7 @@ pub fn garage_features() -> Option<&'static [&'static str]> {
 }
 
 pub fn init_version(version: &'static str) {
-	VERSION.store(Arc::new(version));
+	VERSION.store(Some(Arc::new(version)));
 }
 
 pub fn init_features(features: &'static [&'static str]) {

src/web/Cargo.toml

@@ -14,10 +14,10 @@ path = "lib.rs"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-garage_api = { version = "0.8.2", path = "../api" }
-garage_model = { version = "0.8.2", path = "../model" }
-garage_util = { version = "0.8.2", path = "../util" }
-garage_table = { version = "0.8.2", path = "../table" }
+garage_api.workspace = true
+garage_model.workspace = true
+garage_util.workspace = true
+garage_table.workspace = true
 
 err-derive = "0.3"
 tracing = "0.1"

src/web/web_server.rs

@@ -106,7 +106,7 @@ impl WebServer {
 		addr: SocketAddr,
 	) -> Result<Response<Body>, Infallible> {
 		if let Ok(forwarded_for_ip_addr) =
-			forwarded_headers::handle_forwarded_for_headers(&req.headers())
+			forwarded_headers::handle_forwarded_for_headers(req.headers())
 		{
 			info!(
 				"{} (via {}) {} {}",