gal/constructionbot
Watch
2
0
Fork
You've already forked constructionbot
 0
Code Issues 8 Pull requests Projects Releases Packages Wiki Activity Actions

CI deploy failed on the restart-policy fix push (app unaffected) — diagnose runner failure, land the policy #13

New issue
Open
opened 2026-09-05 17:00:12 +00:00 by jsutter · 0 comments
jsutter commented 2026-09-05 17:00:12 +00:00
Owner
Copy link

2026-09-05: the --restart always policy-fix push (f136104) triggered a CI deploy that FAILED — the app itself is unaffected (the container was manually restarted earlier that day after the Aug-31 kill, currently Up/healthy; the failure happened before the stop/rm phase, so nothing was taken down).

Context: the container had run with restart policy no since its creation and was OOM/killed Aug 31 21:09, sitting down 5 days — the policy fix prevents recurrence once a deploy lands it.

Asks:

  1. Diagnose the failed deploy run (the runner-side failure — logs are not readable via this forge's API; the failure predates the container stop, so likely checkout/build/early step).
  2. Land a green deploy so the --restart always policy takes effect (until then, a tenant restart or another kill leaves the app down again).
  3. Consider the overlay-ghost hardening pattern for this deploy script as well (deepnet/infra#28 second failure mode — see ild/data-governance#2 for the implemented shape).
**2026-09-05:** the `--restart always` policy-fix push (f136104) triggered a CI deploy that FAILED — the app itself is unaffected (the container was manually restarted earlier that day after the Aug-31 kill, currently Up/healthy; the failure happened before the stop/rm phase, so nothing was taken down). **Context:** the container had run with restart policy `no` since its creation and was OOM/killed Aug 31 21:09, sitting down 5 days — the policy fix prevents recurrence once a deploy lands it. **Asks:** 1. Diagnose the failed deploy run (the runner-side failure — logs are not readable via this forge's API; the failure predates the container stop, so likely checkout/build/early step). 2. Land a green deploy so the `--restart always` policy takes effect (until then, a tenant restart or another kill leaves the app down again). 3. Consider the overlay-ghost hardening pattern for this deploy script as well (deepnet/infra#28 second failure mode — see ild/data-governance#2 for the implemented shape).
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
v0.1.2
v0.1.1
v0.1.0
Labels
Clear labels
No items
No labels
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
gal/constructionbot#13
No description provided.