GNOME Bugzilla – Bug 341083
Crash during startup
Last modified: 2006-05-17 08:47:03 UTC
Distribution: Fedora Core release 5 (Bordeaux) Package: Connector Severity: Normal Version: GNOME2.14.1 unspecified Gnome-Distributor: Red Hat, Inc Synopsis: Crash during startup Bugzilla-Product: Connector Bugzilla-Component: Connector Bugzilla-Version: unspecified BugBuddy-GnomeVersion: 2.0 (2.14.1) Description: Description of the crash: When I start Evolition I see it connect to Exchage and start itrs routine of verifying folders. It will run well for a while and then crash Steps to reproduce the crash: 1. Stop evolution 2. Kill all processess associated with evolution 3. restart evolution Expected Results: It would not crash :) How often does this happen? Just about all of the time. Additional Information: I am not sure what is pertaient here... I have a lot of folders (between 90 and 100). I also subscribe to the Fedora list and will often have over 500 messages in that folder. I use filters to move messages to various folders. I do not use virtual folders I thread my messages I have tried with local copy turned on and off. It failes in both situations but more often when turned on. Debugging Information: Backtrace was generated from '/usr/libexec/evolution-exchange' (no debugging symbols found) Using host libthread_db library "/lib/libthread_db.so.1". (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1208858960 (LWP 4340)] [New Thread -1210958944 (LWP 4341)] 0x009f6410 in ?? ()
+ Trace 68101
------- Bug created by bug-buddy at 2006-05-08 21:23 -------
I have turned off the "Automaticall synchronize local folders" option and it appears to be much more stable. After rough 6 to 10 failed restartes I turned the option off and it appears to be coming up this time. I also installed the debuginfo module for the evolution connector. The backtrace that is produced now looks almost identical to what is above. So, unless asked to, I will not include new backtace information. Here is some more version information based on the backtrace: $ rpm -q --whatprovides libc.so.6 glibc-2.4-4 $ rpm -q --whatprovides libthread_db.so.1 glibc-2.4-4 $ rpm -q --whatprovides libglib-2.0.so.0 glib2-2.10.2-1.fc5.1 $ rpm -q --whatprovides libORBit-2.so.0 ORBit2-2.14.0-1 $ rpm -q --whatprovides libpthread.so.0 glibc-2.4-4 $ rpm -qa | grep evolution evolution-2.6.1-1.fc5.2 evolution-connector-2.6.1-1.fc5.1 evolution-connector-debuginfo-2.6.1-1.fc5.1 evolution-data-server-debuginfo-1.6.1-1.fc5.2 evolution-webcal-2.4.1-3.2 evolution-debuginfo-2.6.1-1.fc5.2 evolution-data-server-1.6.1-1.fc5.2
gdb traces are not showing anything useful.
run evolutio-exchange-storage with gdb and then paste better stack trace of crash. If possible paste E2K_DEBUG=4 traces, if u don't want to paste here you can mail it to Sushma. These 90 to 100 folders are folders in exchange server ? Around how many mails you have in your folder. I will try to create a setup like this to replicate this consistent crash.
Please give me the exact gdb command line you wish for me to issue and any specific instructions for collecting the backtrace information I might need. I am a competent programmer so you don't have to spell it out in vast detail but am not gdb savy and would rather get you want you are asking for on the first try instead of the seventh. Yes, all of these folders are on the exchange server. The number of emails in each folder varies greatly. Some are empty and others can have over 500. I would say that on a bell curve that most folders are in the 12-24 emails range. And, if it makes any difference, the exchange server is running Exchange 2000.
Created attachment 65137 [details] Complete exchange folder list. I captured a list of my folders and have uploaded it so that the structure can be seen and reproduced, if needed. This is a screen capture of an OWA session showing all of the folders expanded. If someone knows how to get this information in the form of a text file, that may be more usable.
Thanks for information. type gdb /opt/gnome/lib/evolution/2.6/evolution-exchange-storage. gdb> r then when segmentation fault appears type gdb> thread apply all bt. Then paste stack traces of crash.
Created attachment 65262 [details] E2K_DEBUG=4 output from gdb
Here is the backtrace portion of the dgb output. I have also attacged the complete output from gdb. (evolution-exchange:31017): GLib-GObject-WARNING **: instance of invalid non-instantiatable type `account-removed' [New Thread -1221289056 (LWP 31089)] [Thread -1221289056 (LWP 31089) exited] [New Thread -1221289056 (LWP 31090)] [New Thread -1243616352 (LWP 31092)] [Thread -1221289056 (LWP 31090) exited] [Thread -1210799200 (LWP 31075) exited] [Thread -1233126496 (LWP 31083) exited] [Thread -1243616352 (LWP 31092) exited] [New Thread -1243616352 (LWP 31094)] [Thread -1243616352 (LWP 31094) exited] Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1208428880 (LWP 31017)] 0x0086c0fa in soup_message_io_stop () from /usr/lib/libsoup-2.2.so.8 (gdb) thread apply all bt
+ Trace 68170
Thread 1 (Thread -1208428880 (LWP 31017))
The program is running. Exit anyway? (y or n)
Which is your libsoup version? Using the latest stable libsoup might help. *** This bug has been marked as a duplicate of 322901 ***