ci: verify the images whose healthcheck probe changed - #43
Conversation
#41 switched hris and car-sharing from `wget` to `curl --http2-prior-knowledge`, which also meant swapping the package installed in the runtime stage. Nothing built those images, so a probe that names a tool the image does not carry would have shipped: Docker reports "command not found" exactly like a sick service, so the container just goes unhealthy with no hint why. The new job builds each image and asserts two things about it -- that `curl` is actually present, and that the baked-in HEALTHCHECK speaks HTTP/2. The second assertion is the one that matters: an h2c server cannot be probed over HTTP/1.1 at all, and a silent revert to `wget` would otherwise pass every existing check. Verified both ways against a real image: the current probe passes the guard, a `wget` probe fails it. These two examples need a database to run, so they are built and inspected rather than started. The `with-events-*` images are left out deliberately: they keep `allowHTTP1` at its default of `true`, so their plain `curl -f` probe is correct.
github-actions
Bot
added
the
type:chore
|
Warning Review limit reached
Next review available in: 44 minutes You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthroughThe container E2E workflow now monitors the HRIS and car-sharing Dockerfiles. A matrix job builds both images, verifies ChangesContainer CI coverage
Estimated code review effort: 2 (Simple) | ~10 minutes Possibly related PRs
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches 💡 1🛠️ Fix failing CI checks 💡
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
`pnpm/action-setup` failed with "No pnpm version is specified": it resolves the version from a `packageManager` field, and no example in this repository declares one -- unlike the connectum monorepo, where the root package.json does. npm is the better fit anyway. The `allowBuilds` list in `pnpm-workspace.yaml` exists only because pnpm blocks postinstall scripts, and the one this needs is `@bufbuild/buf` downloading its binary; npm runs it by default. Verified on a clean checkout of hris: `npm install` then `npm run buf:generate` produces `gen/`.
1 participant
Why
#41 switched
hrisandcar-sharingfromwgettocurl --http2-prior-knowledge, which also meant swapping the package installed in the runtime stage. Nothing built those images, so a probe naming a tool the image does not carry would have shipped — and Docker reports "command not found" exactly like a sick service, so the container just goes unhealthy with no hint why.That gap was created by #41, so it is closed here rather than left as follow-up.
What it checks
Builds each image, then asserts two things about the artifact:
curlis actually present —docker run --entrypoint curl … --version.wgetwould otherwise pass every existing check.Verified both directions against a real image:
curl -fsS --http2-prior-knowledge …(current)wget -q --spider …(a revert)gen/is gitignored and the Dockerfiles copy it in, so the job runsbuf:generatefirst.Scope
These two examples need a database to run, so they are built and inspected rather than started — enough to catch the failure mode this change introduced, without standing up Postgres and Temporal.
The
with-events-*images are left out deliberately, not overlooked: they keepallowHTTP1at its default oftrue, so their plaincurl -fprobe is correct for HTTP/1.1 and was never affected.Summary by CodeRabbit