You are not logged in.
I'm getting an error in my dmesg once every boot concerning the journal on my root device:
[ 304.383885] EXT4-fs (sda3): error count: 2
[ 304.383897] EXT4-fs (sda3): initial error at 1402036456: ext4_journal_check_start:56
[ 304.383904] EXT4-fs (sda3): last error at 1402036459: ext4_journal_check_start:56
It doesn't show up immediately upon booting, but shortly after. I assume some kind of journal flush or similar is happening there. I have tried booting from a live USB stick and running fsck, but it comes up clean.
The device is a Samsung 840 SSD, connected via SATA, with the defaults,noatime,discard mount options. SMART shows no errors, apart from ~800 CRC checksum errors, but that value hasn't changed in over 6 months since I changed the SATA cable, which I assume was defective.
Is the journal on my root device corrupt?
E: I tried fsck'ing again, after studying the man page. Re-running fsck and forcing it to check everything even if the filesystem seemed clean, caused a bunch of inode errors to pop up. Studying the files in lost+found afterwards revealed a bunch of source code files that looked like Perl. It must have been a package upgrade that got corrupted during the file transfer from /var to / (they're on separate devices).
Last edited by KozmoNaut (2014-06-16 22:14:58)
Offline