Getting a mild shock because my #forgejo instance suddenly only gave me a "502 - Bad Gateway". Frantically checking log files, because forgejo never crashes. Remembering something. Ah, yeah. That weekly cronjob that stops the container, makes a backup and starts the container again after the backup is done. Yep. Moved that to run at a time where I am most likely AFK (Away From Keyboard). Can sleep now :)
Conversation
Notices
-
Embed this notice
Jan Wildeboer 😷:krulorange: (jwildeboer@social.wildeboer.net)'s status on Friday, 15-Aug-2025 09:10:45 JST
Jan Wildeboer 😷:krulorange:
-
Embed this notice
LovesTha🥧 (lovestha@floss.social)'s status on Friday, 15-Aug-2025 09:13:37 JST
LovesTha🥧
@jwildeboer I just don't trust race conditions about asking aps to have a consistent state and taking snapshots. Being confident the backups actually work when doing them live takes a lot more testing than I'm doing for self hosted things.
-
Embed this notice
Jan Wildeboer 😷:krulorange: (jwildeboer@social.wildeboer.net)'s status on Friday, 15-Aug-2025 09:13:38 JST
Jan Wildeboer 😷:krulorange:
(No need to tell me how to set up backups without stopping the container, I know how to do that. I simply prefer to do it this way for reasons, deal with it ;)
-
Embed this notice