I'm running Debian 11.5 for several months now with no issues until today. No upgrades hardware or software of any kind lately.
All of a sudden services (many of them, logind, hostnamed, colord, atd, etc...) are failing to start. I can't get to a terminal (except for in emergency mode, even a few services fail there).
I think the issue is related to journaling, and journalctl --verify is telling me about file corruption detected etc. etc...
I thought it might be the drive so I changed journal config Storage=volatile and wiped the contents of /var/log/journal/ and restart. Still getting errors. journalctl --verify does not complain about anything on /var but now says:
failed to determine temp directory: not a directory
file corruption detected at /run/log/journal/8bf...etc/system.journal:000000 (of 8388608 bytes 0%)
fail /run/log/journal/8bf...etc/system.journal not a directory
(i can't copy paste from that machine of course so i didn't type out the entire path, it's some 20 to 30 characters long).
My understanding was that /run is tempfs and runs in memory so I swapped out the ram just in case and restarted/tried again. Same errors. I have no idea at this point how to proceed. Any advice/help?
(btw I've got this machine on dual boot and the other OSs don't complain about anything, just this particular Bullseye install. maybe doesn't prove it's not hardware but i doubt it)
Scheduled Maintenance: We are aware of an issue with Google, AOL, and Yahoo services as email providers which are blocking new registrations. We are trying to fix the issue and we have several internal and external support tickets in process to resolve the issue. Please see: viewtopic.php?t=158230
file corruption detected at /run/log/journal
- FreewheelinFrank
- Global Moderator
- Posts: 2107
- Joined: 2010-06-07 16:59
- Has thanked: 38 times
- Been thanked: 230 times
Re: file corruption detected at /run/log/journal
Is your file system full (for the Bullseye partition)?
Output of
?
Try
https://www.debugpoint.com/systemd-journald-clean/
Output of
Code: Select all
$ df -ha
Try
Code: Select all
sudo journalctl --flush --rotate
Code: Select all
sudo journalctl --vacuum-time=1s
-
- Posts: 2
- Joined: 2022-12-04 18:28
Re: file corruption detected at /run/log/journal
None of the partitions are close to full. Varies from 20% to 50% full (I have separate partitions for mounting /, /var, /home) and they're large partitions/drives.
I tried the flush, rotate and vacuum commands you posted above and there was no change after reboot. Still throwing up the corruption errors and still no command line when regular booting (not emergency mode).
I tried the flush, rotate and vacuum commands you posted above and there was no change after reboot. Still throwing up the corruption errors and still no command line when regular booting (not emergency mode).
- FreewheelinFrank
- Global Moderator
- Posts: 2107
- Joined: 2010-06-07 16:59
- Has thanked: 38 times
- Been thanked: 230 times
Re: file corruption detected at /run/log/journal
My guess would be that the corrupt journal files are not the cause but a symptom of a problem, possibly a hardware issue which can cause issues like this in a system that was previously working OK as hardware begins to fail. As you have ruled out RAM, my next suspect would be the HD. The fact that you have another OS booting from that disc does not rule out problems with the disk within the Bullseye partition, causing read write errors and corrupt files.arcticwasteland wrote: ↑2022-12-05 00:27 None of the partitions are close to full. Varies from 20% to 50% full (I have separate partitions for mounting /, /var, /home) and they're large partitions/drives.
I tried the flush, rotate and vacuum commands you posted above and there was no change after reboot. Still throwing up the corruption errors and still no command line when regular booting (not emergency mode).
I would check SMART information for the disk and schedule a check of the file system. You could do one or both of these booting from your alternative OS. I doubt Windows can check a Linux file system (haven't dual booted for about 13 years), in which case you could check the disk from a live CD or USB.