Documents the second runner and the job order
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
@@ -930,3 +930,15 @@ Verder heb ik de vraag ook beantwoord"
|
||||
**Context**: OPERATIONS - Deployment pipeline adjustment (during Monitoring Setup approval wait)
|
||||
|
||||
---
|
||||
|
||||
## Deployment - Runner Tests
|
||||
|
||||
**Timestamp**: 2026-10-19T12:58:55Z
|
||||
|
||||
**User Input**: "In theorie kan de raspberry pi de lighthouse nog wel runnen toch? De raspberry pi runne ris namelijk de hoofdrunner die altijd aan staat en alles moet kunnen runnnen anmelijk"
|
||||
|
||||
**AI Response**: "Confirmed: the Pi has every label the workflow uses (linux for check/lighthouse, raspberry-pi for deploys) with the same job image as before. Test results: run 538 (PR, 0eec9a4) failed: lighthouse on the PC runner, chrome-launcher treated the container as WSL (is-wsl sees 'microsoft' in the shared WSL kernel; is-docker 2.2.1 only checks /.dockerenv, which Podman does not create) and tried /mnt/undefined/Users/undefined/AppData/Local. Fixed in the PC runner config (empty /.dockerenv via --volume /dev/null:/.dockerenv:ro), not in the project; also fixed start-runner.ps1 (PowerShell 'if (podman container exists)' tested output, not exit code, so the old container was never replaced). Run 539 (dispatch, both runners): check Pi, deploy-test Pi in parallel with lighthouse on the PC, all green. Run 540 (dispatch, PC runner stopped): Pi ran check, then deploy-test, then lighthouse, all green. PC runner restarted. Updated deployment-instructions 1.1 and deployment-plan D1. Docs committed locally, not pushed (a push starts another full run); commit/push permission for this change ends here."
|
||||
|
||||
**Context**: OPERATIONS - Deployment pipeline adjustment verified
|
||||
|
||||
---
|
||||
|
||||
Reference in New Issue
Block a user