Bug 275420

Summary:	Sometimes Kontact and dbus-daemon hog two of my 6 cores with 100% each - killing Kontact solves this
Product:	[Applications] kontact	Reporter:	Ekkehard Blanz <Ekkehard.Blanz>
Component:	general	Assignee:	kdepim bugs <kdepim-bugs>
Status:	RESOLVED UNMAINTAINED
Severity:	normal	CC:	Ekkehard.Blanz, kdebugzilla01, kdenis, michael
Priority:	NOR
Version:	4.4.10
Target Milestone:	---
Platform:	Ubuntu
OS:	Linux
Latest Commit:		Version Fixed In:
Sentry Crash Report:

Description Ekkehard Blanz 2011-06-11 19:30:19 UTC

Version:           4.4.10 (using KDE 4.6.2) 
OS:                Linux

When Kontact runs for more than a day on my otherwise very stable Kubuntu 11.04 64-bit system, it causes dbus-daemon and itself to hog all the CPU cycles they can get.  Two of my 6 cores are completely pegged at 100%.  When I then kill Kontact, the problem is temporarily solved.  Unfortunately, starting Kontact again will bring me right back to that situation.  Sometimes Kontact won't even start again while dbus-daemon hogs 100% of a core.  I have to restart to be able to run Kontact normally again.  Please let me know if there is any other information I should provide the next time it does that.

Reproducible: Sometimes

Steps to Reproduce:
Let sit Kontact for some time - often happens when I come back to the computer in the morning.

Actual Results:  
Two CPU cores are pegged at 100% - would render a single core machine useless.

Expected Results:  
Kontact and dbus-daemon behaving "nice"ly, using less than solid 100% CPU.

OS: Linux (x86_64) release 2.6.38-8-generic
Compiler: cc

Comment 1 Jan Drewes 2011-06-15 10:38:44 UTC

Similar for me. Every now and then (up to several times a week), dbus-daemon hogs a cpu-core, and when the system is in that state, no applications will launch anymore.
Killing just akonadiserver will help, but only for a limited time; after a short while, dbus-daemon will hog the CPU again, at which point only restarting seems to help reliably. 
Killing & restarting dbus-daemon doesn't help because KDE apps don't seem to find the new daemon afterwards. 

Can provide additional infos as needed...

Comment 2 Christophe Marin 2011-06-16 23:18:55 UTC

> Version:           4.4.10 (using KDE 4.6.2)

Please retry with Kontact 4.6.0

Comment 3 michael 2011-08-24 18:42:19 UTC

Application: kontact (4.4.7)
KDE Platform Version: 4.4.5 (KDE 4.4.5)
Qt Version: 4.6.3
Operating System: Linux 2.6.32-trunk-686 i686
Distribution: Debian GNU/Linux 6.0.2 (squeeze)

-- Information about the crash:
<In detail, tell us what you were doing  when the application crashed.>

The crash can be reproduced every time.

 -- Backtrace:
A useful backtrace could not be generated

Report to https://bugs.kde.org

Just to note, this may not be related to this bug, but I use Kontact with about 10 IMAP accounts, and some of those are shared IMAP accounts, and every 2 days I  have to close kmail/kontact and then manually kill all the processes.. about 6-20 processes hand around, and kontact also hangs around.  From the GUI, if you make the mistake of trying to start kontact, it appears to try to start, then dies silently, no warning that another kontact instance is still running.  The IMAP processes seem to be stuck on: select(4, [3], [], NULL, NULL, while the kontact process is hung with:

read(8, 0x8268d68, 4096)                = -1 EAGAIN (Resource temporarily unavailable)

and it loops.. It also seems that the reason we get the system CPU runaway, is that the kontact process keeps spawning new IMAP connections, but doesn't close the legacy ones that it lost contact with or is waiting on..

Comment 4 Denis Kurz 2017-06-24 00:16:29 UTC

Closing due to lack of feedback