Timeline for Cannot umount systemd mounted luks container in raid
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 21, 2024 at 15:51 | comment | added | telcoM | I repeat: take a backup ASAP. Your first disk has already died, and the second is about to die. The inability to complete an unmount indicates the OS has cached data it would need to write, but the remaining disk is no longer accepting any writes - at the firmware level. And that means the fsck command would also fail to make any changes to the disk. | |
| Dec 21, 2024 at 15:39 | comment | added | Oleksa | Sadly, one disk is failed, the other one is OK. I do not have a possibility to travel to the server and would like to fix the filesystem remotely without rebooting the server. So, the main question is how to unmount the ext4 file system completely to run fsck, if this file system resides in a luks container which resides in a logic volume. | |
| Dec 21, 2024 at 15:00 | comment | added | telcoM | Check the status of the RAID1 set. The symptoms would match the situation where one disk of the RAID1 pair is already completely dead and the other disk has started failing too. If that's your situation, make a backup of all important data on that RAID set before doing anything else. It is quite possible that a filesystem check cannot fix this problem, and an actual hardware repair/replacement is needed. | |
| Dec 21, 2024 at 13:41 | history | asked | Oleksa | CC BY-SA 4.0 |