GNOME Bugzilla – Bug 366403
crash in Evolution: Getting mail from 2 acco...
Last modified: 2006-11-04 03:29:15 UTC
What were you doing when the application crashed? Getting mail from 2 accounts: one had asked me for its new password to login and get messages; the other was fetching over 5000 messages from the server: this is not the first time I have logged in to get mail from here: I've "fetched mail" before, and aborted before all the messages came down because it takes a long time. And so, actually, this time I cancelled again; and that was when this error popped up. Distribution: Ubuntu 6.10 (edgy) Gnome Release: 2.16.1 2006-10-02 (Ubuntu) BugBuddy Version: 2.16.0 Memory status: size: 368508928 vsize: 0 resident: 368508928 share: 0 rss: 188551168 rss_rlim: 0 CPU usage: start_time: 1162055793 rtime: 0 utime: 3896 stime: 0 cutime:2820 cstime: 0 timeout: 1076 it_real_value: 0 frequency: 0 Backtrace was generated from '/usr/bin/evolution-2.8' (no debugging symbols found) Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1233176912 (LWP 5367)] [New Thread -1326748768 (LWP 5383)] [New Thread -1318356064 (LWP 5382)] [New Thread -1309484128 (LWP 5380)] [New Thread -1286612064 (LWP 5379)] [New Thread -1278219360 (LWP 5378)] [New Thread -1269826656 (LWP 5376)] [New Thread -1260278880 (LWP 5372)] [New Thread -1251886176 (LWP 5371)] (no debugging symbols found) 0xffffe410 in __kernel_vsyscall ()
+ Trace 79371
*** Bug 366404 has been marked as a duplicate of this bug. ***
Thanks for the bug report. Unfortunately, that stack trace is not very useful in determining the cause of the crash. Could you please install some debugging packages [1] and reproduce the crash, if possible? Once bug-buddy pops up, you can find the stacktrace in the "details", now containing way more information. Please copy that stacktrace and paste it as a comment here. Thanks! [1] debugging packages for evolution, evolution-data-server and gtkhtml, plus debugging packages for some basic GNOME libs. More details can be found here: http://live.gnome.org/GettingTraces/DistroSpecificInstructions
Hi. I have not seen another crash. I am installing a couple of debug packages, so hopefully if it does crash I'll see something there. However, Evolution definitely has some severe problems on my system. It definitely seems to have a memory leak. The system monitor shows it using 2.7Gig of virtual memory. It' been running for a couple of hours now. I so happen to have about 6300 messages in the trash. Not sure if the large number of messages is contributing. ZOIKS! As if on cue, it just crashed in the background. That has been typical: most crashes seem to happen while its running in the background. Hopefully not a paging bug or something... I'll add another comment with the stack trace.
Traces: per above: hopefully one of the debug packages I did get to install is useful... Distribution: Ubuntu 6.10 (edgy) Gnome Release: 2.16.1 2006-10-02 (Ubuntu) BugBuddy Version: 2.16.0 Memory status: size: -1330122752 vsize: 0 resident: -1330122752 share: 0 rss: 594817024 rss_rlim: 0 CPU usage: start_time: 1162498400 rtime: 0 utime: 6929 stime: 0 cutime:5412 cstime: 0 timeout: 1517 it_real_value: 0 frequency: 0 Backtrace was generated from '/usr/bin/evolution-2.8' Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread -1232951632 (LWP 4848)] [New Thread -1318163552 (LWP 4866)] [New Thread -1309770848 (LWP 4865)] [New Thread -1300898912 (LWP 4863)] [New Thread -1292506208 (LWP 4862)] [New Thread -1284113504 (LWP 4861)] [New Thread -1261438048 (LWP 4857)] [New Thread -1251664992 (LWP 4854)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 82107
Thread 7 (Thread -1261438048 (LWP 4857))
Hi. It appears by looking at the system monitor that each time Evolution sends/receives mail it claims more memory and never releases that. When it gets mail from my server, most times, Evolution says it's "expunging messages" for a few moments and then completes normally. -- I have that account set to "leave messages on the sever", and "delete them when 120 days old". I used to have that at 365 days and there were almost 8000 messages on my server; but now it's down to about 2500 (I can see by checking with a webmail tool instead of evolution). But it doesn't say it's expunging messages every single time. In any case, I can reliably chew up all 3Gigs of virtual memory by just clicking "Send/Receive" enough times. And this behavior began when I installed Edgy: it didn't exhibit any of these things in the previous release. If you know in particular another debug package I should install, let me know. I am afraid I can't get it to reliably crash, but maybe by chewing up all the VM I can trigger it. Tx.
Here's another crash. I instigated it by setting my account to check mail every 4 minutes; and I manually hit send/receive frequently. Eventually it chewed up all the VM (at least 2.7Gigs of the 3Gig total, not including the few other apps I'm running in the meantime [like NetBeans]). And sure enough, bug buddy popped up eventually. I managed to install one more debug package. So here's the stack trace: Distribution: Ubuntu 6.10 (edgy) Gnome Release: 2.16.1 2006-10-02 (Ubuntu) BugBuddy Version: 2.16.0 Memory status: size: -1383882752 vsize: 0 resident: -1383882752 share: 0 rss: 662937600 rss_rlim: 0 CPU usage: start_time: 1162570697 rtime: 0 utime: 6469 stime: 0 cutime:4987 cstime: 0 timeout: 1482 it_real_value: 0 frequency: 0 Backtrace was generated from '/usr/bin/evolution-2.8' Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread -1233197392 (LWP 5712)] [New Thread -1373635680 (LWP 5741)] [New Thread -1310401632 (LWP 5729)] [New Thread -1302008928 (LWP 5728)] [New Thread -1292727392 (LWP 5726)] [New Thread -1261438048 (LWP 5725)] [New Thread -1269830752 (LWP 5719)] [New Thread -1251910752 (LWP 5716)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 82432
Thread 7 (Thread -1269830752 (LWP 5719))
Great, nicve traces, thanks Steven! Looking at the stacktrace, one additional debugging package may be interesting, though I'm not sure -- libglib-2.0, which is in the top most parts of these traces. If you could install this package and reproduce again, please either paste the resulting trace as a comment to bug 352284, or attach it to that bug. Thanks. Following up on bug 352284. *** This bug has been marked as a duplicate of 352284 ***