[systemd-devel] Heavy IO -- need help on how to debug

[Kok, Auke-jan H]

On Sat, Nov 10, 2012 at 6:13 AM, Pedro Francisco
<pedrogfrancisco at gmail.com> wrote:
> Yesterday my system was slow and I checked iotop: systemd-journal was
> using 99% IO (but not CPU). I haven't restarted the machine, but kill
> systemd-journal didn't help, the process would respawn -- as expected
> -- and resume the heavy IO).
>
> I disabled storage ("Storage=none") and restarted the service. It
> helped a lot, but I still get the ocasional slow down -- though a lot
> less severe.
>
> iotop output (idle prio is because I did a ionice -c3 on it):
> 23616 idle root        3.14 M/s   14.86 K/s  0.00 % 99.62 % systemd-journald
>
> $ df -h
> Filesystem      Size  Used Avail Use% Mounted on
> rootfs           15G   11G  4,1G  72% /
>
> System: up-to-date Fedora 17, x86
> $ sudo rpm -qa systemd*
> systemd-sysv-44-21.fc17.i686
> systemd-44-21.fc17.i686
> systemd-analyze-44-21.fc17.i686
>
> What else can I do to debug what is going on next time it happens?

I'm suspecting that the old version has many bugs that are currently
solved, so I would just suggest getting the latest version or building
from git first.

In general, sysprof would help you to narrow down what parts of the
code are consuming a lot of time. After that, you'd run a debugger
attached to the journal.

Cheers,

Auke