On Monday, August 3, 2020 12:42:13 PM EDT Robbie Harwood wrote:
> On Saturday, August 1, 2020 1:27:07 PM EDT Steven Grubb wrote:
>> I was using my desktop system when I got logged out. After logging back
>> in, I found this message in my logs:
>>
>> Aug 1 13:08:22 x2 journal[1751]: UID 1000 exceeded its 'bytes' quota
on
>> UID 1000.
>
> I wrote a script that searched every binary on my system to see what
> possibly matches this output. Turns out this cryptic message is from
> dbus-broker. As best I can tell, a KDE program is triggering this. And I
> have no idea how to reconfigure things to fix it, but the failure is
> catestrophic when it shouldn't be. And its happening with some
> regularity.
>
> You are warned...
When reporting problems of this kind, could you share a link to the
bugzilla you've filed against dbus-broker
https://github.com/bus1/dbus-broker/issues/236
Still discussing this...
or this KDE program?
No bug filed yet. I need to review my logs carefully to see if I can pinpoint
something relevant. But akonadiconsole seems to be part of what triggers this
catastrophic failure.
But this leads into another topic. akonadiconsole segfaults. So does akonadi.
They both call Dr Konqui. It asks if you want to file a report. You say yes.
Then it does something and says no backtrace generated, can't file a report.
So, that leads me to wonder, why do we enable Dr Konqui which cannot do
anything and not let abrtd process it so that all the segfaulting KDE stuff
gets collected up somewhere actionable instead of being lost forever? Not
having any visbility into crashes occuring in KDE components is not helping
improve software quality.
-Steve