0

I have a Dell Edge 5000 gateway that has Ubuntu 18.04 installed on it. It runs fine, but it fails to boot every 15 to 20 boots. There are several error messages that are displayed that all seem related.

systemd-journald[250]: Failed to rotate /var/log/journal/XXXXXXXXXXXXXXXXXXXXXXX.journal: Read-only file system

systemd-journald[250]: Failed to write entry (X items, XXX bytes), ignoring: Input/Output error

blk_update_request: I/O error, dev sda, sector XXXXXXXX op 0x0: (READ) flags 0x0 phys_seg 1 prio class 0

It will display a combination of these messages and pop up with a few more lines every few seconds. This will keep up indefinitely until I cycle power.

I have checked the disk as many ways as I can think of. I have used the BIOS scan tool (long and short scans), fsck, smartctl (including long scan), and badblocks. Everything passes and shows no errors.

Please let me know if you have any ideas. I'm convinced that something else is causing the drive to mount read only. I'd rather not specify in fstab to ignore filesystem errors.

I appreciate the help.

EDIT:

I learned more today.

If run "fsck -Cf /dev/sda2" to force a scan of the drive, everything checks out just fine.

However, if I run "fsck -Cfc /dev/sda2", to force the badblocks scan, I have have problems. It does report any bad blocks, but it will fail when checking inodes. It will say that every single inode is bad despite having passed in the previous command. It gets even more interesting if I just reboot. The reboot will cause the BIOS to not recognize the hardware. If I power down, the next startup will fail. The next power cycle will boot fine.

I also caught where the failure was occurring on boot today, but I dont know how helpful it is. I have not had a chance to fully exhaust google yet.

EXT4-fs error (device sda2): __ext4_find_entry:1532: inode #1638541: comm gdm3:reading directory iblock 0

It just went into the same journal errors after this. I'll keep this updated as I find out more. Thanks for the help so far.

Garrett Anderson
  • 1
  • 1
    These really are usually hardware errors, and if not the drive itself, could be the connection. Have you tried re-seating the drive in the machine? – popey Dec 05 '20 at 20:25
  • There's a similar question here which makes a good case for updating your SSD firmware. Also checking for BIOS updates at the same time – Simon B Dec 05 '20 at 20:37
  • That was a good post. I did verify that the firmware was up to date per the Apacer website. I also updated BIOS to see if that would fix it. Still similar problems. I have not tried reseating the drive yet. The computer is still under warranty with Dell. I make sure I'm not going to void anything by doing that, but I'll be calling them in the morning. Thanks for the responses. – Garrett Anderson Dec 07 '20 at 02:35

1 Answers1

0

It ended up being a bad hard drive. Replaced it, and everything started working fine.

Garrett Anderson
  • 1