Problem restarting Nextcloud after Disk Full Event

[jdaviescoates]

@Shai said in Problem restarting Nextcloud after Disk Full Event:

In terms of replacing config/config.php with a version from a backup, the most recent version I had was from 2023

I don't understand, you're using Cloudron but don't have regular automatic backups?!?

[Shai]

Yes, really stupid, there were reasons which I don't want to go into here.

[robi]

search for "disk full nextcloud" in the forum for ready made solutions

[girish]

@Shai take a config.php from a fresh config.php and patch in values like passwordsalt, secret . If you restart nextcloud, values like db/redis/smtp/ldap will get updated to latest values automatically, so you don't need to fixup those.

[Shai]

@girish followed your instruction but I'm getting the following in the logs after a restart of the Nextcloud App says "Not responding":

An unhandled exception has been thrown:
Jan 11 22:12:08OCP\HintException: [0]: Downgrading is not supported and is likely to cause unpredictable issues (from 30.0.4.1 to 29.0.3.4) ()code_text

Ideas?

[girish]

@Shai there is a version field in the same config.php . Adjust that to your nextcloud version.

[Shai]

Thanks @girish. Won't restart after the recommended edit of config.php. Here is the error:

=> Healtheck error: Error: connect ECONNREFUSED 172.18.17.145:80
OCP\HintException: [0]: Downgrading is not supported and is likely to cause unpredictable issues (from 30.0.4 to 29.0.3.4) ()

[girish]

@Shai said in Problem restarting Nextcloud after Disk Full Event:

30.0.4

Is this the version field you filled in?

[Shai]

@girish
Yes.

[girish]

@Shai I think you have to contact us at support@ at this point, if you want us to fix it.

[timconsidine]

Off-topic perhaps but after you have recovered your Nextcloud deployment …

This seems to happen to me every couple of years

Worth considering some kind of free space monitoring?
Various server monitor apps/agents, but simplest is a bash script run as a cron using ntfy to send alerts.
I get a daily notification of disk free space, and an “intra-day” alert if anything spikes over a set level.

I’m not judging, been there too, but poking yourself in the eye repeatedly doesn’t seem a good idea. Incidents happen, I know, the trick is to learn from them. ‘Root cause’ ‘corrective action’ blah blah blah