mirror of
https://github.com/rustfs/rustfs.git
synced 2026-08-02 03:19:19 +00:00
Zhengchao An
b7805caa58
The four consolidated ci keys plus the build keys still do not fit the repository's fixed 10GB Actions cache quota, so LRU keeps evicting them: measured demand is ci-dev 2331MB + ci-feat-proto 2310MB + ci-feat-rio 1951MB + ci-uring 1317MB + two build legs at ~1429MB + cargo-deny 844MB, and main pushes add two more build legs. That is roughly 14.4GB against 10.24GB. The symptom is misleading: Cache Warm reports every job successful and the restores log "full match: true", yet ci-feat-rio and ci-uring disappear from the cache list between runs. cache-all-crates was the wrong default for this repository. With it set to true, rust-cache's cleanup returns before pruning ~/.cargo/registry/src and its config archives the whole registry, so every cache carried the unpacked source tree of every dependency — not, as the name suggests, just a few extra crates. Setting it to false is rust-cache's own default and loses no coverage: the package set comes from `cargo metadata --all-features`, a strict superset of any single lane's feature closure; -sys crates are explicitly exempt from pruning, since their source timestamps would otherwise trigger rebuilds; and everything pruned is re-unpacked from the .crate files still in registry/cache, whose mtimes crates.io normalises, so cargo fingerprints stay valid. Applied to the setup composite and to audit.yml's own rust-cache. Cache Warm now also reports the sizes of registry/src, registry/cache, registry/index, ~/.cargo/git and target/ to the step summary, immediately before rust-cache's post step archives them, so the size of the effect is measured rather than assumed. The new sizes only appear once the cache key next rotates, since rust-cache skips the save entirely on an exact key hit. Deliberately not forcing that by bumping prefix-key: it would invalidate every family at once and produce a repository-wide cold build. Refs: rustfs/backlog#1598, rustfs/backlog#1600
82 lines
3.0 KiB
YAML
82 lines
3.0 KiB
YAML
# Copyright 2026 RustFS Team
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
# Asserts that the self-hosted runners are still ephemeral — one job per pod.
|
|
#
|
|
# This repository is public and its pull_request jobs run on those runners,
|
|
# executing the PR's own build.rs, proc-macros and tests. The only thing keeping
|
|
# that code from reaching a later job is that each ARC pod handles exactly one
|
|
# job and is then destroyed. That guarantee lives in the ARC scale-set
|
|
# configuration, outside this repository, where it can be changed without any PR
|
|
# — so it is asserted here from the outside, against real run data, instead of
|
|
# being assumed.
|
|
#
|
|
# Monthly rather than per-PR: the property changes only when someone
|
|
# reconfigures the scale set, and the check costs a few dozen API calls.
|
|
# See docs/ci/runners.md and rustfs/backlog#1602.
|
|
|
|
name: Runner Hygiene
|
|
|
|
on:
|
|
schedule:
|
|
- cron: "0 6 1 * *" # Monthly, 1st at 06:00 UTC (after the daily audit cron)
|
|
workflow_dispatch:
|
|
|
|
permissions:
|
|
contents: read
|
|
|
|
concurrency:
|
|
group: runner-hygiene
|
|
cancel-in-progress: false
|
|
|
|
jobs:
|
|
check-ephemerality:
|
|
name: Check runner ephemerality
|
|
runs-on: ubuntu-latest
|
|
timeout-minutes: 15
|
|
steps:
|
|
- name: Checkout repository
|
|
uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
|
|
with:
|
|
persist-credentials: false
|
|
|
|
# Exit 2 (inconclusive / broken) is deliberately not a pass: a window
|
|
# where every sm-* job was still queued would otherwise look identical to
|
|
# a clean bill of health.
|
|
- name: Assert one job per self-hosted runner
|
|
env:
|
|
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
|
|
run: ./scripts/ci/check_runner_ephemerality.sh 40
|
|
|
|
alert-on-failure:
|
|
name: Alert on scheduled failure
|
|
needs: [check-ephemerality]
|
|
# Same ci-8 mechanism as coverage.yml, audit.yml and the nightly lanes:
|
|
# scheduled runs file a tracking issue, manual dispatch stays quiet so
|
|
# debugging never produces a spurious alert.
|
|
if: always() && github.event_name == 'schedule' && contains(needs.*.result, 'failure')
|
|
runs-on: ubuntu-latest
|
|
timeout-minutes: 10
|
|
permissions:
|
|
contents: read
|
|
issues: write
|
|
steps:
|
|
- uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7
|
|
with:
|
|
persist-credentials: false
|
|
- name: Open or update failure-tracking issue
|
|
uses: ./.github/actions/schedule-failure-issue
|
|
with:
|
|
github-token: ${{ secrets.GITHUB_TOKEN }}
|