josh/electric-cherry
1
0
Fork 0
Code Issues 18 Pull Requests Actions Packages Projects Releases Wiki Activity

Rebuild missing containers during maintenance recovery

Browse Source
This commit is contained in:
josh
2026-10-08 13:33:01 -04:00
parent 368e457dc6
commit b3fa486886
9 changed files with 201 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/guides/maintenance.md
+12 -6
View File
@@ -127,15 +127,21 @@ an application restart, routine watchdog checks wait for a new status-monitor pa
**Check now** and recovery verification still request fresh evidence, and
Provisioning retains its checks before executing recovery actions.
Three observations over at least two minutes are required before an automatic
health restart; starting or restarting containers get five minutes. Unknown host-agent responses are observations,
not proof of a stopped container. Repeated manual checks do not accelerate this
Three observations over at least two minutes are required before automatic
health recovery; starting or restarting containers get five minutes. A confirmed
missing container goes directly to rebuild after this same sustained detection
window. It cannot contain players, so recovery skips the player-count check.
Offline hosts, unknown container state, and missing or stale observations wait
for fresh evidence. Repeated manual checks do not accelerate this
window. A successful routine or manual game-status probe between scans resets
the failure streak. Missing RCON configuration, unresolved ports, and invalid
replies remain diagnostic findings and cannot authorize health recovery.
replies alone remain diagnostic findings and cannot authorize health recovery.
A health finding gets one restart, then one rebuild if recovery fails after five
minutes and fresh health checks still justify recovery. If current container and
A health finding for an existing container gets one restart, then one rebuild if
recovery fails after five minutes and fresh health checks still justify recovery.
Missing-container recovery gets one direct rebuild. Both paths respect maintenance
policy and pause settings, intentionally stopped servers, capacity limits, and
protected beta or campaign work. If current container and
game checks succeed before an automatic health restart or rebuild is submitted,
the obsolete operation is cancelled and the finding is resolved. Unknown health
holds recovery for another check. Drift repairs and requested lifecycle operations