You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Lazurio upstream watch has not run on its schedule since 2026-11-06 #190
The scheduled Lazurio upstream watch workflow has not run since 2026-11-06, so upstream's stable v0.1.9 opened no rebuild comment. The rebuild bot's daily check caught it instead (#185, decision N).
Evidence (2026-11-12 07:40 UTC)
gh run list -R Lazurio/t3code --workflow lazurio-upstream-watch.yml lists daily schedule runs up to 2026-11-06T13:12:36Z, all success, and none after.
gh api repos/Lazurio/t3code/actions/workflows reports the workflow active.
On main (cab18ad99b) the trigger is unchanged: schedule: cron "41 5 * * *" and workflow_dispatch. The file last changed in b92f7c6a7f (2026-11-05).
main was replaced by the v0.1.7 rebuild on 2026-11-06 (cab18ad99b), the same day as the last scheduled run.
Upstream published v0.1.9 on 2026-11-11T11:33:49Z. The run due on 2026-11-12 05:41 UTC would have commented on the open rebuild issue.
Cause: not verified. The timing matches the replacement of main, but nothing here shows that GitHub dropped the schedule because of it.
Next step for someone who may change the repository's workflows: check the workflow's schedule in the Actions settings, and run it once with workflow_dispatch to see whether it still works. If replacing main drops schedules, the swap step in docs/operations/lazurio-fork-release.md ("Výměna main") should say how to bring this one back, and the rebuild skill should check it after each swap.
Until then the rebuild bot's daily check covers new upstream stable releases.
What stopped: the last schedule run is 42223647750 (2026-11-06 13:12:36Z, success). Before that the schedule ran every day; GitHub created the runs between 10:56 and 13:49 UTC (cron 41 5 * * *).
What happened around it:
The v0.1.7 swap of main landed 2026-11-06 18:40:37Z. Its push run of Lazurio Fork CI, 42272856043, is still queued today: the repository's Actions were off at that point.
From 2026-11-06 18:40Z to 2026-11-08 11:23Z no run of any workflow was created. Actions were turned back on on 2026-11-08; since then push and workflow_dispatch runs work, but no schedule run arrived, although the workflow reported active.
The force push of 2026-11-05 21:46Z was followed by normal scheduled runs on 11-05 and 11-06, so a rewrite of main alone does not stop the schedule. My reading (not verified on GitHub's side): the schedule was dropped while Actions were off and turning them on again did not register it.
Today gh api repos/Lazurio/t3code/actions/permissions returns "enabled": true.
Restore (2026-11-13, with the Organization Admin's permission for this one workflow):
gh workflow disable 384066175 → disabled_manually at 17:59:01Z, gh workflow enable 384066175 → active at 17:59:08Z.
One manual run: 43478153474success, "main is built on v0.1.9; upstream's newest stable release is v0.1.9."
Not verified yet: whether the schedule is back. That needs the scheduled run due 2026-11-14 05:41 UTC (expected to be created several hours late, like before). If it does not arrive by 2026-11-15 05:41 UTC, the next step is a commit on main that changes the workflow file, which registers the schedule again. Until then the release bot's daily check covers new upstream releases.
Runbook and skill:#194 adds the post-swap check (Fork CI really started; the next day, a scheduled watch run arrived) and the restore steps.
Activity
immakermatty commentedon Nov 13, 2026
Cause and restore (2026-11-13)
What stopped: the last
schedulerun is 42223647750 (2026-11-06 13:12:36Z, success). Before that the schedule ran every day; GitHub created the runs between 10:56 and 13:49 UTC (cron41 5 * * *).What happened around it:
mainlanded 2026-11-06 18:40:37Z. Its push run ofLazurio Fork CI, 42272856043, is stillqueuedtoday: the repository's Actions were off at that point.pushandworkflow_dispatchruns work, but noschedulerun arrived, although the workflow reportedactive.mainalone does not stop the schedule. My reading (not verified on GitHub's side): the schedule was dropped while Actions were off and turning them on again did not register it.gh api repos/Lazurio/t3code/actions/permissionsreturns"enabled": true.Restore (2026-11-13, with the Organization Admin's permission for this one workflow):
gh workflow disable 384066175→disabled_manuallyat 17:59:01Z,gh workflow enable 384066175→activeat 17:59:08Z.success, "main is built on v0.1.9; upstream's newest stable release is v0.1.9."Not verified yet: whether the schedule is back. That needs the scheduled run due 2026-11-14 05:41 UTC (expected to be created several hours late, like before). If it does not arrive by 2026-11-15 05:41 UTC, the next step is a commit on
mainthat changes the workflow file, which registers the schedule again. Until then the release bot's daily check covers new upstream releases.Runbook and skill: #194 adds the post-swap check (Fork CI really started; the next day, a scheduled watch run arrived) and the restore steps.
Agent: environment a480495c-25e0-4e51-828b-1dbf1b77d62f · bot c59bfd74-62a6-4eaa-88dd-5e19f1b3032e
docs(lazurio): check the upstream watch schedule after the main swap
docs(lazurio): check the upstream watch schedule after the main swap
docs(lazurio): check the upstream watch schedule after the main swap
docs(lazurio): check the upstream watch schedule after the main swap
docs(lazurio): check the upstream watch schedule after the main swap