(In reply to Кощеев from comment #50)
> pcp version 5.2.0
>
> $ find /usr/lib/systemd/system /etc/systemd/system -type f | xargs grep
> pmlogger_check.path
> grep: /usr/lib/systemd/system/system-systemdx2dcryptsetup.slice: Нет такого
> файла или каталога
well that didnt find anything, but we
Hi all,
The current release of PCP for f32 is pcp-5.2.0-1.fc32.x86_64.rpm
https://dl.fedoraproject.org/pub/fedora/linux/updates/32/Everything/x86_64/Packages/p/pcp-5.2.0-1.fc32.x86_64.rpm
contains all of the PCP service units and none of them reference
pmlogger_check.path.
So exactly what
Fernando,
the logs you sent me indicate pmlogger is starting but exiting very soon
after (and then being repeatedly restarted by systemd). I don't see any
evidence of a systemd unit configuration problem. The PCP archives are
being created, but have a corrupted label record and I can't read any
(In reply to Fernando Viñan-Cano from comment #40)
> I'm now on 5.1.0-1.fc31 and this problem still exists and just fills up the
> log with these messages constantly.
>
> Is this ever going to be fixed?
Hi Fernando, it should be fixed now. Perhaps after your upgrade you need
to run systemctl
Closed as per the fix from Comment #30 and verified with Marko's recipe from
Comment #21.
The additional work for changing the pmlogger service to use type=notify is a
separate issue being tracked in upstream issue
https://github.com/performancecopilot/pcp/issues/798
--
You received this bug
(In reply to Marko Myllynen from comment #32)
> I would also support a new PCP upstream release to address this or at least
> a new updated build for the affected distributions with selected fixes
> included (instead of waiting over a month for the next scheduled PCP
> release). Thanks.
agree we
Committed an upstream patch that fixes the service timeout reproducer test
posted by Marko in Comment#21, along with a bunch of QA updates.
commit 0f3a1fa7d0d7a898acf2ee4a6f23b626d4fb7362
Author: Mark Goodwin
Date: Wed Nov 13 09:50:57 2019 +1100
logutil: fix pmlogger service timeout on
Hi Fernando, I can repro this issue on a virgin install (beginning with
no PCP installed at all). This fails to start the pmlogger service with
a timeout - as reported. If I then: systemctl restart pmlogger.service
it seems to work OK and we end up with the pmlogger service running. Are
you seeing
