Arch Linux

wolflarson

Member

Registered: 2014-03-22

Posts: 3

systemd-journal process using 100% of one cpu core after reboot

journalctl -b showed
"Process 2016 (virtuoso-t) dumped core."

more than a few times and virtuoso-t would was using more cpu than it does most of the time.

at 14 minutes of uptime systemd-journal stopped using 100% and went back to normal. Anyone know a way to figure out what happened here?

update:

 systemd-coredump[3182]: Process 3133 (virtuoso-t) dumped core.
 systemd-journal[5407]: Permanent journal is using 2.6G (max allowed 4.0G, trying to leave 4.0G free of 50.7G available → current limit 4.0G).
 systemd[1]: systemd-journald.service watchdog timeout!
 systemd[1]: Starting Journal Service...
 systemd[1]: systemd-journald.service stop-sigterm timed out. Killing.
 systemd[1]: Starting Journal Service...
 systemd[1]: systemd-journald.service: main process exited, code=killed, status=9/KILL
 systemd[1]: Unit systemd-journald.service entered failed state.
 systemd[1]: systemd-journald.service has no holdoff time, scheduling restart.
 systemd[1]: Stopping Journal Service...
 systemd[1]: Starting Journal Service...
 systemd[1]: Started Journal Service.
 systemd-journal[5407]: Journal started
 dbus[365]: [system] Activating via systemd: service name='org.freedesktop.ModemManager1' unit='dbus-org.freedesktop.ModemManager1.service'
 dbus[365]: [system] Activation via systemd failed for unit 'dbus-org.freedesktop.ModemManager1.service': Unit dbus-org.freedesktop.ModemManager1.service failed to load: No such file or directory.
 systemd-coredump[3289]: Process 3235 (virtuoso-t) dumped core.
 

it happened again here is what the journal was saying while it was happening (i removed lots of (virtuoso-t) dumped core that happened ... 10 or more times)

Thanks in advance,
Wolf Larson

Last edited by wolflarson (2014-03-22 15:36:37)

WonderWoofy

Member

From: Los Gatos, CA

Registered: 2012-05-19

Posts: 8,414

Re: systemd-journal process using 100% of one cpu core after reboot

A few things about posting....

First of all, please use code tags when you paste things like log snippets and whatnot.  You can see how to do this by following the 'BBCode' link below.  There are a number of other tricks that you should learn to use (they're pertty simple) on that page as well.

Second, there is not a lot of context as to what is going on here.  You have basically said that it appears that the journal is taking more CPU than it should and that you have a program dumping its core.  These two things are different.   It appears as though you have cherry picked journal messages and posted them here, which doesn't give a very good idea of what is happening overall in the machine during this time.

And most importantly (not about posting)...

Welcome to the Forums!

wolflarson

Member

Registered: 2014-03-22

Posts: 3

Re: systemd-journal process using 100% of one cpu core after reboot

well from what i am reading in the log journal (and I really have no idea if i am interpreting this correctly) it seems like journal is getting to big because it it being continuously written to and then it times out and gets restarted then my issues stop. how can i find out what caused this?

ps.
if you would like i can pastebin the whole journal but most of it wont help i would think. That is the part that was being outputted when I was having issues. would you like the whole thing?

WonderWoofy

Member

From: Los Gatos, CA

Registered: 2012-05-19

Posts: 8,414

Re: systemd-journal process using 100% of one cpu core after reboot

In the Newbie Corner of the forums, there are a few stickied threads that are recommended reads.  One of which is this one.

If the logs are not too too much, then it can be posted here with code tags. If it is crazy big, pastebin it and then post the link.

If this is really all the information it gives about the journal being nuts, then you might want to enable debug levels of logging for the journal and then see if you can gather more information that way.  This can either by done by using a drop-in snippet to the unit file, or by turning on debug levels for the entire system from the kernel command line.

The journal is not designed in such a way that it can be continously written to.  There are safeguards in place to ensure that masively spamming services are cut off after a certain number of log messages in a given timeframe.  Additionally, even the flush from the journal in RAM to the disk is only done periodically.

wolflarson

Member

Registered: 2014-03-22

Posts: 3

Re: systemd-journal process using 100% of one cpu core after reboot

per http://forum.kde.org/viewtopic.php?t=92886 I have disabled Nepomuk. Thanks WonderWoofy for the link i will try to follow those guidelines in the future when doing any forum posts.

if the issues does not pop up again how do i mark this solved?

fsckd

Forum Fellow

Registered: 2009-06-15

Posts: 4,173

Re: systemd-journal process using 100% of one cpu core after reboot

if the issues does not pop up again how do i mark this solved?

Edit your first post and add [SOLVED] to the title as described in here.

---

aur S & M :: forum rules :: Community Ethos
Resources for Women, POC, LGBT*, and allies