fix(site-deploy): drop job-level concurrency — deadlocks callers holding their own group - #51
Merged
Merged
Conversation
…hat hold their own group Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The first live deploy through
site-deploy.yml@v1.1.0failed in 0 seconds with:Mechanism: the reusable's job hardcoded
concurrency: { group: deploy-main }. Every site caller holds a workflow-leveldeploy-maingroup of its own (correctly — a caller serializing its deploys is caller policy). The caller's workflow takes the lock, the called job then waits on the same group name, and GitHub detects the parent-child deadlock and cancels the run. (site-lentago-dev run 32057853117; the site kept serving — ECS never rolls unless a deploy succeeds — but the deploy pipe on main was dead until this fix.)Fix: remove the job-level concurrency from the reusable entirely, with a comment explaining why it must not come back. Serialization policy belongs to the caller; a library workflow imposing its own lock can only collide with it.
Also bumps the internal
render-claude-summaryrefs to v1.1.1 per RELEASING.md step 2, since this commit is the v1.1.1 tag target.actionlintpasses on the changed workflow.Process note for the record: this bug was invisible to every static check — actionlint clean, inputs validated, tag resolved — and appeared only on the first real invocation. It is exactly why the caller migration is staged one site at a time: the blast radius was one repo's deploy pipe for a few minutes, not three.