[systemd-devel] Antw: [EXT] Re: Odd status after core dump

[Ulrich Windl]

>>> Lennart Poettering <lennart at poettering.net> schrieb am 04.09.2020 um 17:08
in
Nachricht <20200904150837.GG267847 at gardel-login>:
> On Mo, 03.08.20 08:32, Ulrich Windl (Ulrich.Windl at rz.uni‑regensburg.de)
wrote:
> 
>> OK, some additional notes (maybe these issues are fixed in a newer version
>> already):
>> Looking in my /var/log/messages, I see higher resolution of time stamps,
and
>> specifically I see the original core‑dump message:
> 
> Use "journalctl ‑o short‑precise" to get µs granularity.
> 
>> 2020‑07‑31T11:31:57.404198+02:00 v04 systemd[1]: iotwatch at NFS1.service:
Main
>> process exited, code=dumped, status=6/ABRT
>> 2020‑07‑31T11:31:57.421639+02:00 v04 systemd[1]: Stopped iotwatch I/O
>> performance monitor instance "NFS1".
>>
>> (Using "journalctl I see lower‑resolution timestamps as output)
>> Jul 31 11:31:57 v04 systemd[1]: iotwatch at NFS1.service: Main process
exited,
>> code=dumped, status=6/ABRT
>> Jul 31 11:31:57 v04 systemd[1]: Stopped iotwatch I/O performance monitor
>> instance "NFS1".
>> Jul 31 11:31:57 v04 systemd[1]: iotwatch at NFS1.service: Unit entered failed
>> state.
>> Jul 31 11:31:57 v04 systemd[1]: iotwatch at NFS1.service: Failed with result
>> 'core‑dump'.
>>
>> So why is the "core dump complete" included in the status of "starting"
the
>> service when the core dump happened when stopping the service, and why
isn't
>> the original core dump message included as well?
> 
> I cannot parse this, and I cannot tell you how you configured your
> syslog and why it shows less data.

Actually I think it's easy to parse and the config is plain SLES12 SP5.

> 
> Lennart
> 
> ‑‑
> Lennart Poettering, Berlin