Repository navigation
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Upstream PR creation is restricted, so this review PR is hosted in the fork. Its base matches
google/axmain.A missing gateway currently turns into an allow-all policy. Record a failed Ready/GatewayReady condition and stop startup on lookup errors. Suspension remains available: it runs before policy application, so a missing gateway cannot prevent stopping an actor or cause a wildcard policy during suspension.
Fixes google#350.
TDD and validation
This branch starts independently from
origin/main(d8ed0fe) and has two commits:41e0e71adds the regression and fails on the original behavior.fa56783implements the fix; the regression and complete suite pass.Reproduce the red/green transition by checking out either commit and running:
Validated:
go test ./... -count=1, targeted-racetests,go mod tidywith unchanged module files,make build, andgit diff --check. Controller tests use an in-process Substrate fake; no deployed cluster was used.Review
Codex review: addressed the suspension regression identified in the first pass; second pass found no actionable regressions. Some reviewer test runs were blocked by its listener sandbox; the validation above was run separately with local listener access.