Mailinglist Archive: opensuse-bugs (3094 mails)

< Previous Next >
[Bug 931964] systemd: BUG: soft lockup - CPU#0 stuck
  • From: bugzilla_noreply@xxxxxxxxxx
  • Date: Mon, 08 Jun 2015 13:28:55 +0000
  • Message-id: <bug-931964-21960-nIn44SbqVI@http.bugzilla.suse.com/>
http://bugzilla.suse.com/show_bug.cgi?id=931964

--- Comment #15 from Thomas Blume <thomas.blume@xxxxxxxx> ---
(In reply to Roland Bernet from comment #14)
As it's not too likely I'm just sitting in front of the terminal/screen when
the machine locks up, should I change the boot options to something like

systemd.log_level=debug systemd.log_target=syslog debug

Would I still need a big log_buf_len ?

The system is completely on local disks, only home and data disks are
mounted over NFS. Therefore I'm not sure there is any log writing to NFS as
you mentioned earlier.

Yes ok, just try it.
My goal is to find out what is behind the systemd pid that causes the CPU hang.
In the latest hang it was:

PID: 3097 Comm: systemd

Instead of verbose systemd log, maybe you can just create a script that runs:

systemctl status

in a loop and writes the output to some log file.
When the system hangs, we can hopefully find the pid in this file.
But please be aware that systemctl status is quite heavy in terms of system
resource usage.
So, don't let it loop on a machine where performance is important.

Alternatively, you could try to trigger a kernel crashdump via sysrq after the
machine hung. This should also contain the information.

--
You are receiving this mail because:
You are on the CC list for the bug.
< Previous Next >