Bound how long Upload Assets can hang - #543
Merged
Merged
Conversation
A full sync of all 128 graphics takes ~3 minutes, but nothing capped the job. Today a Configure AWS credentials step stalled on connect ETIMEDOUT to sts.us-east-2.amazonaws.com; it ran 19 minutes before being cancelled by hand, with hours of headroom left under GitHub's 6-hour default. Two bounds, because they fail differently: timeout-minutes: 15 on the job — a backstop for a hang in any step, not just this one. Five times the observed full-sync duration. action-timeout-s: 120 on the credentials step — the action defaults to no timeout and 12 retries, and a TCP connect timeout costs ~6m45s per attempt, so the default budget is ~81 minutes. Retries are kept, just lowered to 3: transient STS errors are real and worth retrying, but the OIDC token expires partway through a long retry loop, so later attempts cannot succeed anyway. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
tashian
enabled auto-merge
July 27, 2026 22:07
dopey
approved these changes
Jul 27, 2026
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Follow-up to #541 / #542. Today's
full_syncrun stalled and I cancelled it by hand at 19 minutes. Nothing in the workflow would have stopped it before GitHub's 6-hour default job timeout.What happened
3.146.15.71issts.us-east-2.amazonaws.com. Each attempt burns ~6m45s in the TCP connect timeout — the action's own backoff is trivial (12ms, 58ms) — so 12 attempts is a ~81 minute budget for a failure that cannot succeed, because the OIDC token expires partway through.Two bounds
A backstop for a hang in any step, not just this one. A full sync of all 128 graphics takes ~3.2 minutes, so this is ~5x observed.
The action defaults to no timeout and 12 retries. Retries are worth keeping — transient STS errors are real — but 12 is pointless when the token dies first. 120s caps the step inside the first failed attempt.