evolution memory leaking, system thrashing

[Tobias Pflug]

Binary package hint: evolution

(intrepid with all updates installed running on a SamsungQ45)

Several times I have been experiencing massive harddisk usage / swap usage
and the system got so unresponsive that I couldn't log in on any tty as
the login attempts would time out after 60 seconds.

This happens at seemingly random times without me doing any
special actions.

When I used sysrq / oom_kill (out of memory killer) it killed evolution
and shortly afterwards everything was back to normal.

I have attached the part of /var/log/messages containing SysRq
information gathered during the incident. Please let me know
if there is more information that I can provide to help.

[Sebastien Bacher (seb128)]

Thank you for taking the time to report this bug and helping to make Ubuntu better. Please try to obtain a valgrind log following the instructions at https://wiki.ubuntu.com/Valgrind and attach the file to the bug report. This will greatly help us in tracking down your problem.

[Michele Costantino Soccio (michelinux)]

I don't know whether it's related, but I noticed that the evolution-data-server-2.24 process eats a lot of memory: usually it's betweeb 300MB and 400MB of resident memory. This is definitely too much even for modern machines.

[Pedro Villavicencio (pedro)]

Closing this bug report as no further information has been provided. Please feel free to reopen this bug if you can provide the information asked for. Thanks!.

[Reece (reece)]

This problem is very real. Please reopen.

I'm running evo 2.24 on Intrepid on a HP NW8440. I have the above symptoms daily.

I've attached valgrind output, generated thus:
$ evolution --force-shutdown
$ sleep 3
$ G_SLICE=always-malloc G_DEBUG=gc-friendly valgrind -v --tool=memcheck --leak-check=full --num-callers=40 --log-file=/home/rkh/evo-valgrind/2008-11-13-14:30:09.valgrind evolution --component=mail

I've not used valgrind and am unsure that this completed correctly. In my case evo RSS goes to 1.8G or so, then is killed. I don't know whether the valgrind output is complete in such cases.

-Reece

[Sebastien Bacher (seb128)]

the log has no useful information, how did you get it? could you send the bug to bugzilla.gnome.org?

[Reece (reece)]

My bad. I was so sure that I'd installed the dbg symbols on this machine that I didn't actually look. I was wrong. I'm running w/valgrind now. I will report back here and to b.g.o).

[Sebastien Bacher (seb128)]

did you get a new log and opened the bug to GNOME?

[Reece (reece)]

Not yet. This was happening many times per day over several days, but seems to have stopped. I've used suspend/resume with wireless a lot less lately, and I had already suspected that involvement in the Evo issues. Evo still crashes 1-2 times/day, but I have no data.

In any case, let's close this until I can provide some useful evidence for either problem.

-Reece

[Martin Mai (mrkanister-deactivatedaccount-deactivatedaccount)]

We are closing this bug report because it lacks the information we need to investigate the problem, as described in the previous comments. Please reopen it if you can give us the missing information, and don't hesitate to submit bug reports in the future. To reopen the bug report you can click on the current status, under the Status column, and change the Status back to "New". Thanks again!