This server release improves reliability for CI runs on Tuist-managed runners.
- Recover stuck jobs automatically. If a workflow job is assigned to a runner but the runner never registers with GitHub, the server now detects the mismatch and re-queues the job so another runner can pick it up. Jobs no longer sit idle in GitHub’s queue for hours when a runner fails to start.
- Fix runner bootstrap failures. Resolved an infrastructure issue that could trap runner hosts in a retry loop while loading firewall rules, so newly deployed runners can start successfully.