Repository navigation
Restore Codex source acquisition and reliable CI image preparation - #271
Merged
Merged
Conversation
HTExplicit
enabled auto-merge (squash)
October 9, 2026 21:13
HTExplicit
changed the title
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Source acquisition reused the target STATE-probe builder, including short probe instructions, encrypted-output flags, synthetic session/window headers and forced connection close. The pinned ranxi source path instead uses an ordinary OAuth account-test request. Live tests passed on an initial candidate but later acquisition cycles still produced rejected routes; this source-contract discrepancy remained after target validation was corrected.
Build source acquisition from the existing ordinary account-test payload and normal source connection policy. Keep its configured account identity, credentials, proxy, medium effort, bounded lifetime and observation isolation. Target two-shot probes retain their independent full-probe contract. No account settings, network routes, protocol switches or STATE acceptance conditions are changed.
Validation: source payload/header/connection contract, original source proxy/lifetime behavior, stream terminal handling, bounded candidate replacement and target probe identity/tier regressions are run narrowly. Actual cross-lease behavior remains pending deployment and is not claimed fixed from code checks alone.
Required integration checks were blocked by Docker Hub anonymous pull limits and registry timeouts. CI now preloads the identical PostgreSQL, Redis and Ryuk OCI digests from publisher-operated ECR/GHCR registries, then applies the existing local image names. The mirror digests were verified against Docker Hub tag metadata. Test definitions, database versions and required checks are unchanged; preparation fails immediately if a fixed image cannot be pulled.
The candidate builder also hit Docker Hub when bootstrapping BuildKit. Candidate and release builds now use the same BuildKit digest via Google's public cache and a BuildKit registry mirror for unchanged Dockerfile references. Builder, Dockerfile frontend and all four base-image tag digests were verified equal to Docker Hub metadata. Existing fixed-version publication and production deployment guards remain in place.