Hi,
I started to see the same bug with the generic kernel. I believe it started 
with 'linux-image-6.8.0-84-generic' but it took me a bit to notice and track 
down, so it may have been the case sooner as the logs rotated since. I see it 
happening on two different hardware machines with kernels

  * 6.8.0-86-generic
  * 6.8.0-84-generic

Both machines are 'Ubuntu 24.04.3 LTS' and they have ZFS
`2.2.2-0ubuntu9.4`. The zfs is only used by LXD for VMs/containers, it
is not on root.

The bug is triggered (I believe) by following script:

  echo 3 >/proc/sys/vm/drop_caches
  echo 1 >/proc/sys/vm/compact_memory

which runs daily by cron. I also suspect that this is linked to other
activities that are triggered by cron at the same time because when I
execute the script manually, at arbitrary time, it doesn't reproduce the
crash. Though I haven't figured out what exactly is the contributing
factor.

Both machines have sufficient RAM. They have 16GB and the usual workload
leaves about 10GB free. I also run monitoring (by prometheus) and I
don't see any suspicious spikes, though the scrape interval is 1m, so it
might miss sudden spike before crash.

I can also attach apport report if it helps.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2069025

Title:
  kernel BUG at mm/migrate.c:663 on 6.8.0-1009-aws #9-Ubuntu

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-signed-aws/+bug/2069025/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to