signing + tools fetch: retry the transient release-edge class - #169
Merged
Merged
Conversation
The 0.16.24 publish lost a musl signing leg to an HTTP 500 on a manifest download — after every asset had already signed and converged — and #166's windows leg to an intermediary 403. Both clear on a re-ask; both cost a human round-trip instead. The executor seam (every gh call in the signer rides it) now re-asks the 5xx / intermediary-403 class with bounded backoff — deterministic failures (404, auth, usage) still raise at once — and the workflow's one direct gh release download gets the same budget in bash. Specs pin the retry contract: success on re-ask, no budget spent on 404, the full budget named on a persistent transient.
ronaldtse
force-pushed
the
fix/sign-download-transients
branch
from
September 16, 2026 11:16
39ca505 to
6ccfb41
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Two of the 0.16.24 publish's failures were GitHub release-edge transients under load, each costing a rerun cycle:
gh release download … manifest.json→ HTTP 500, after all five targets had already signed and converged.ListArtifacts→ intermediary 403 (the sibling matrix leg downloaded the same artifact fine).upload_release.rbalready carries a ~4.5-minute retry budget; downloads had none.scripts/sign_release.rb: the executor seam re-asks the5xx/ intermediary-403class (bounded 4 attempts, exponential backoff); deterministic failures raise immediately with the same NAMED FAILURE shape._build-platform.yml: the one directgh release download(windows tfs CLI fetch) gets the same budget in bash.