Repository navigation
Bound code swap to ten-task services and remove admission - #228
Merged
Merged
Conversation
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]>
…ailures Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]>
The integrated gate observed a backward WSL wall-clock adjustment during the elapsed-time assertion. Keep the same backoff threshold using performance.now(). Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Remove the allocator function, stream, slots, grants, reclamation and 500/501 pressure fixtures. Preserve checked startup, task-stop budgets, ECS registration and candidate-only tracing. Qualify the cap with ten-task lifecycle and ordinary image fallback fixtures. Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]>
matuscongrady
changed the title
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Check live surge during the existing fleet preflights, and overlap RoleId, release-head and local bundle reads. All prerequisites still succeed before prepare, and activation still waits for the pushed image. Co-Authored-By: Claude Opus 5.5 <[email protected]>
Give the benchmark app its own default port, support a focused over-cap image transition, and include CLI status reads in the regional cost envelope. Co-Authored-By: Claude Opus 5.5 <[email protected]>
Keep the fresh IAM lookup and identity checks, but overlap them with independent metadata reads instead of delaying image construction. Co-Authored-By: Claude Opus 5.5 <[email protected]>
Refuse unknown scheduled/controller bounds and count overridden autoscaling maxima before installing or preparing a supervisor. Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]>
Co-Authored-By: Claude Opus 5.5 <[email protected]> # Conflicts: # scripts/workspace/test-registry.ts
matuscongrady
marked this pull request as ready for review
October 19, 2026 20:59
Co-Authored-By: Claude Opus 5.5 <[email protected]>
# Conflicts: # apps/console
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
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.
Code swap now uses normal images when a service can exceed ten tasks, including its autoscaling maximum and deployment surge. The CLI checks before opt-in and on each deploy, accounts for service/scaling overrides, and prints one reason when it chooses images. A static 500-poller stack budget includes one draining generation. The regional model estimates $7.44/month for ten tasks under the documented envelope.
This removes the capacity Lambda, stream, admission grants, slots, reclamation and 500/501 fixtures after the owner narrowed the feature to small dev/staging fleets. Readiness, task-stop budgets, ECS registration, narrowed ptrace and the #214 waiter remain. Fresh identity and fleet checks still run; independent reads overlap with metadata/image work.
Live verification passed: ten-task scale-out and replacement loaded committed code before readiness, with zero errors across 2,050 requests. Raising the maximum selected an ordinary image, printed the cap reason and served the new code. The 20-edit regression passed at p50 4.768 s, p95 11.905 s and max 15.737 s, with zero failures across 7,811 timed and 2,252 continuous requests, no fallback/replacement and qualified AWS clock fits. Earlier misses and fixture failures remain in the evidence ledger.
Validation: 19 focused tests, pinned Go race/vet/build, built runtime smoke and integrated checks passed. The actual template has zero cfn-lint errors and nine existing warnings. All seven current runs are cleaned up, with independent checks confirming stacks, tables, buckets, ECR, log groups, Lambdas and every exact ECS revision are absent. This is source-only qualification; protected customer activation remains in later phases.
Stack order: #226 → #227 → #228. The Console launch-note companion console-app#90 must land before #228. Its CI job could not start because GitHub reports an account billing/spending-limit block; local integrated and documentation checks passed.
🤖 Generated with Claude Code