Restart instances that crash after running for 10 minutes

Adds instance.restart_after_crash (default off) and
instance.crash_retry_attempts (default 3, 0 for no limit). The host only
restarts an instance that ran for at least 10 minutes before exiting
unexpectedly, so a boot loop stops after one crash. A user stop, kill or
restart resets the count; it is not persisted across host restarts.

Closes #1251

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
bbassieandClaude Opus 5.5 committed 2026-11-21 15:55:28 +00:00
1 parent 7a2aade25b
commit 565e931a96
5 files changed
+142 -1

No files matched your search

+17
View File
@@ -549,6 +549,23 @@ Does not affect start or stop of the instance while the host is running.
Defaults to false.
### instance.restart_after_crash
If enabled the host starts this instance again when the Factorio server exits unexpectedly after running for at least 10 minutes.
A crash sooner than that is logged as an error and the instance is left stopped, as it is likely to crash again.
Defaults to false.
### instance.crash_retry_attempts
Number of times the host restarts this instance after a crash before giving up and logging an error.
The count resets when the instance is stopped or restarted by a user, or the host is restarted.
Set to 0 for no limit.
Defaults to 3.
### instance.exclude_from_start_all
If enabled this instance will be excluded from the "Start all" button operation in the web interface.