You are not logged in.

#1 2021-01-11 20:04:24

EnochRoot
Member
From: Just outside London, UK
Registered: 2015-12-08
Posts: 43

Hard disk timeouts and other issues at reboot

Have been seeing some strange issues at reboot in my Debian testing installation. Was seeing messages about systemd timing out waiting for hard disks to become available.

Jan 03 19:14:16 hostname systemd[1]: Timed out waiting for device /dev/disk/by-uuid/UUID.This was happening for the mount of all non root partitions.

The system would drop into emergency mode, running sudo mount -a would successfully mount all the hard disks and boot could then continue, sometimes successfully, and other times with lightdm not responding to keyboard or mouse events.

I did some research and people with a similar issue typically had changed their hard disks or reformatted so their system was trying to mount an old UUID. There were suggestions to increase the timeout period to ensure that udev had longer to detect the hard disks. I added the timeout option to the three non root mount points which were timing out. 

x-systemd.device-timeout=300s

This got the machine to boot successfully (though a lot slower than before). However when lightdm was started, it was not responsive to keyboard or mouse events. I logged into the box from my laptop and found that restarting lightdm fixed the issue.

I think this has been happening since I upgraded to the 2nd last kernel in testing, probably 5.9.02 or 5.9.0.3

I get the feeling that my workaround is masking the issue rather than fixing it. Anyone know what might have caused udev / the kernel to take longer to detect the hard disks ? Also systemd seems to be running a fsck on my boot partition on every boot.

Thanks

Enoch

Offline

#2 2021-01-13 17:42:34

ohnonot
...again
Registered: 2015-09-29
Posts: 5,387
Website

Re: Hard disk timeouts and other issues at reboot

Of course you should research Debian testing channels first.
Then:
Basically you have to look at the whole journal and try to find the origin of this cascade of weird behaviour.
Obvious questions: do those partitions exist? Have you fsck'd them to see if maybe the hardware is onthe way out? smartctl?
Have you made changes to systemd's default behaviour?
Show us your fstab. Are these exactly the same partitions systemd runs through at boot?


Please use CODE tags for code.
Search youtube without a browser: repo | thread
BL quote proposals to this thread please.
my repos / my repos

Offline

Board footer

Powered by FluxBB