GNOME Bugzilla – Bug 340778
E-D-S 1.6x crashes talking to Groupwise
Last modified: 2006-07-31 09:07:17 UTC
From: Scott Anderson <ee_in_co@yahoo.com> To: submit@bugs.gnome.org X-Mailer: bug-buddy 2.14.0 Subject: E-D-S 1.6x crashes talking to Groupwise Distribution: Debian testing/unstable Package: Evolution-Data-Server Severity: critical Version: GNOME2.14.1 1.6.x Gnome-Distributor: Debian Synopsis: E-D-S 1.6x crashes talking to Groupwise Bugzilla-Product: Evolution-Data-Server Bugzilla-Component: Calendar Bugzilla-Version: 1.6.x BugBuddy-GnomeVersion: 2.0 (2.14.1) Description: Description of the crash: I'm on Debian unstable, I know the "unstable" is very literal right now. The groupwise backend is crashing very frequently. I'm reporting this bug to see if we can gather some details on why. The Debian developers think the old glib free problems should be gone in the latest version. We'll see. After the crash, I must slay bonobo before I can launch evolution again. Is there a different work-around, because I have to close all my applications before I can relaunch evolution, which is very time consuming. Steps to reproduce the crash: 1. Launch evolution 2. Click on Calendar 3. E-D-S crashes Expected Results: How often does this happen? Every time. Additional Information: Debugging Information: Backtrace was generated from '/usr/libexec/evolution-data-server-1.6' (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 -1489250624 (LWP 19170)] [New Thread -1525924944 (LWP 19204)] [New Thread -1492087888 (LWP 19171)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 68046
Thread 1 (Thread -1489250624 (LWP 19170))
__________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ------- Bug created by bug-buddy at 2006-05-05 21:01 -------
I tried to start E-D-S with GROUPWISE_DEBUG=1 set. It immediately crashed. Here's the stderr: ~> GROUPWISE_DEBUG=1 /usr/lib/evolution/evolution-data-server-1.6 evolution-data-server-Message: Starting server e-data-server-Message: adding type `EBookBackendGroupwiseFactory' e-data-server-Message: adding type `EBookBackendFileFactory' e-data-server-Message: adding type `ECalBackendContactsEventsFactory' e-data-server-Message: adding type `EBookBackendLDAPFactory' e-data-server-Message: adding type `EBookBackendVCFFactory' e-data-server-Message: adding type `ECalBackendGroupwiseTodosFactory' e-data-server-Message: adding type `ECalBackendGroupwiseEventsFactory' e-data-server-Message: adding type `ECalBackendFileTodosFactory' e-data-server-Message: adding type `ECalBackendFileEventsFactory' e-data-server-Message: adding type `ECalBackendFileJournalFactory' e-data-server-Message: adding type `ECalBackendWeatherEventsFactory' e-data-server-Message: adding type `ECalBackendHttpTodosFactory' e-data-server-Message: adding type `ECalBackendHttpEventsFactory' e-data-server-Message: adding type `ECalBackendCalDAVEventsFactory' ** (evolution-data-server:21323): WARNING **: Failed to send buffer ** (evolution-data-server:21323): WARNING **: Failed to send buffer libedata-book-Message: Error registering the PAS factory: generic error evolution-data-server-ERROR **: server.c:372: could not initialize Server service "BOOKS"; terminating aborting...
The warnings show that the EDS libraries might not installed properly. What is the version of EDS your using ? Try to kill the bonobo-activation server and restart EDS. ** (evolution-data-server:21323): WARNING **: Failed to send buffer libedata-book-Message: Error registering the PAS factory: generic error evolution-data-server-ERROR **: server.c:372: could not initialize Server service "BOOKS"; terminating aborting... Means that EDS is not able to start at all. Please correct the environment variable BONOBO_ACTIVATION_PATH. Usually all the bonobo server files would be present in the path <prefix>/lib/bonobo/servers/ . After setting it kill the bonobo-activation-server and try starting EDS.
I have another clue on what is causing the E-D-S crash. I saved the output of E-D-S to a logfile. When I came in to work today, all the backend services to evolution had crashed, and the E-D-S logfile showed this on the last two entries: libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1411: Getting object list ((completed-before? (make-time "20060408T232329Z"))) ***MEMORY-ERROR***: evolution-data-server[3021]: GSlice: assertion failed: sinfo->n_allocated > 0 The memory error does explain why the application will run for a while and then crash when I'm not interacting with it. For further information, here's what I did to follow your previous instructions: So I ran 'bonobo-slay' and it restarted all the apps on my desktop. Then when I launch evolution, it did not immediately crash. I checked with the debian evolution maintainers, and they checked my installation and everything is up-to-date with debian unstable. You can see the details at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=362482. My bonobo server files are at /usr/lib/bonobo/servers/. Do I need to set BONOBO_ACTIVATION_PATH before I start evolution, or somewhere before I log in to Gnome? Does everyone who has evolution or E-D-S crash have to slay bonobo, or do I have something wrong that goes deeper than the typical debian unstable installation?
I discovered that I did not yet have the evolution-data-server-dbg installed. Here's a backtrace from bugbuddy after evolution-data-server crashed today. I hope it has more debugging symbols. Debugging Information: Backtrace was generated from '/usr/libexec/evolution-data-server-1.6' Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread -1489525056 (LWP 7654)] [New Thread -1538405456 (LWP 7921)] [New Thread -1513096272 (LWP 7768)] [New Thread -1492382800 (LWP 7657)] 0xffffe410 in __kernel_vsyscall ()
+ Trace 68154
Thread 1 (Thread -1489525056 (LWP 7654))
I have not mentioned to do a bonobo-slay. killing the activation server will not restart all the apps. And this need to be done every time. Just when you installed evolution and if EDS does not start due to the error, thats when killing the bonobo-actication server helps. (evolution-data-server:21323): WARNING **: Failed to send buffer ** (evolution-data-server:21323): WARNING **: Failed to send buffer libedata-book-Message: Error registering the PAS factory: generic error evolution-data-server-ERROR **: server.c:372: could not initialize Server service "BOOKS"; terminating aborting... The second trace does not have much information. Are you able to see your groupwise calendar items ? Do you get same terminal warnings even now ?
Thanks for the help. I am confused in part about the use of bonobo-slay. Does killing the activiation server mean to run bonobo-slay, or just send the bonobo-activation-server process a SIGINT or SIGKILL using 'kill'? Do I need to kill the activation server every time? The comment above is inconsistent about when that is needed. Now when I run evolution-data-server: ~> GROUPWISE_DEBUG=1 /usr/lib/evolution/evolution-data-server-1.6 > /var/tmp/evolution-data-server.log 2>&1 I get: e-data-server-Message: adding type `ECalBackendGroupwiseTodosFactory' e-data-server-Message: adding type `ECalBackendGroupwiseEventsFactory' e-data-server-Message: adding type `ECalBackendFileTodosFactory' e-data-server-Message: adding type `ECalBackendFileEventsFactory' e-data-server-Message: adding type `ECalBackendFileJournalFactory' e-data-server-Message: adding type `ECalBackendWeatherEventsFactory' e-data-server-Message: adding type `ECalBackendHttpTodosFactory' e-data-server-Message: adding type `ECalBackendHttpEventsFactory' e-data-server-Message: adding type `ECalBackendCalDAVEventsFactory' evolution-data-server-Message: Server up and running So E-D-S starts up fine at first. I can check the Groupwise calendar, add and delete appointments/meetings, etc, *for a while*. But within about 10 minutes, E-D-S crashes and it almost always prints to STDERR/STDOUT: ----------------- libecalbackendgroupwise-Message: ******** found libecalbackendgroupwise-Message: ******** found libecalbackendgroupwise-Message: ******** found libecalbackendgroupwise-Message: ******** found ***MEMORY-ERROR***: evolution-data-server[17766]: GSlice: assertion failed: sinfo->n_allocated > 0 ----------------- I have also had E-D-S crash when reading the Groupwise Frequent Contacts addressbook or updating the Groupwise Tasks. It prints out essentially the same error. ***MEMORY-ERROR***: evolution-data-server[12567]: GSlice: assertion failed: sinfo->n_allocated > 0
(In reply to comment #6) > Thanks for the help. I am confused in part about the use of bonobo-slay. Does > killing the activiation server mean to run bonobo-slay, or just send the > bonobo-activation-server process a SIGINT or SIGKILL using 'kill'? it will just kill the bonobo-activation-server process. > > Do I need to kill the activation server every time? No. Doing it might solve the issue only when you get the warning such as the following ** (evolution-data-server:21323): WARNING **: Failed to send buffer libedata-book-Message: Error registering the PAS factory: generic error evolution-data-server-ERROR **: server.c:372: could not initialize Server service "BOOKS"; terminating aborting... Once the environment variables are set properly, these warning would not appear. > I have also had E-D-S crash when reading the Groupwise Frequent Contacts > addressbook or updating the Groupwise Tasks. It prints out essentially the > same error. Please attach the gdb traces.
(In reply to comment #7) > (In reply to comment #6) > > Thanks for the help. I am confused in part about the use of bonobo-slay. Does > > killing the activiation server mean to run bonobo-slay, or just send the > > bonobo-activation-server process a SIGINT or SIGKILL using 'kill'? > it will just kill the bonobo-activation-server process. > > > > Do I need to kill the activation server every time? > No. Doing it might solve the issue only when you get the warning such as the > following > ** (evolution-data-server:21323): WARNING **: Failed to send buffer > libedata-book-Message: Error registering the PAS factory: generic error > evolution-data-server-ERROR **: server.c:372: could not initialize Server > service "BOOKS"; terminating > aborting... > Once the environment variables are set properly, these warning would not > appear. Are you certain the environment variables are the problem? The Debian maintainers seem to think it has nothing to do with the environment variables. We just had a discussion about the mail-notification program perhaps causing the problem. Since I have removed mail-notification, I have not seen this error again (yet...) > > I have also had E-D-S crash when reading the Groupwise Frequent Contacts > > addressbook or updating the Groupwise Tasks. It prints out essentially the > > same error. > Please attach the gdb traces. Will do shortly...
I cannot get a backtrace it seems. I'll attach the logfile while running E-D-S under gdb. You'll see the 'thread apply all bt' returned nothing after the crash.
Created attachment 66615 [details] GDB running E-D-S (no backtrace avail)
Created attachment 66617 [details] Bug buddy report of evolution crash I had experienced the same type of crash as usual. I was using the Contact Manager, looking at my Frequent Contacts. Then I clicked on my LDAP contact folder. Evolution reported the LDAP folder had terminated abnormally and would not be available until evolution restarted. Then the Groupwise Tasks, Calendar, Frequent Contacts each popped up a similar message. So I quit evolution using the GUI pulldown menu. Evolution crashed and Bug Buddy popped up getting me a stack trace. Let me know if this is useful (I am doubting it).
Created attachment 66625 [details] E-D-S crash composing new mail Another crash, this time E-D-S. I did not have the stdout of E-D-S saved this time. E-D-S was launched from evolution. After using the evolution mailer to read mail, then leaving it idle for about an hour, I came back to send a message. Immediately after clicking the compose new mail icon in the toolbar, bug buddy reported that E-D-S crashed and I saved this backtrace. Let me know if it is useful. I am not certain that the Groupwise interaction caused this crash.
Created attachment 66792 [details] Crash autocompleting addresses composing mail Here's another crash when composing mail.
The traces show that, the crash happens in addressbook. Moving it to addressbook.
hey scott the autocompletion crash is fixed in head. closing it for now, if you still face any crashers, please reopen it against the appropriate component.
(In reply to comment #15) > hey scott the autocompletion crash is fixed in head. > closing it for now, if you still face any crashers, please reopen it against > the appropriate component. > What version of E-D-S has the fix in it? I'm still running 1.6.1 and updating soon to 1.6.2.
I have another crash that is not related to the addressbook. It may have been the Tasks component this time. Following the suggestion to reopen if I face any more crashers, I went looking for more reasons why my evolution crashes. It didn't take more than 1 minute. I launched E-D-S like this: % GROUPWISE_DEBUG=1 /usr/lib/evolution/evolution-data-server-1.6 2>&1 | tee /var/tmp/evolution-data-server.log and evolution like this: % GROUPWISE_DEBUG=1 evolution-2.6 2>&1 | tee ~/.evolution/debug.log I was able to view my Groupwise and local mail for several folders. Then I switched to my calendar view. Immediately an error message popped up that my Tasks had crashed, and immediately following that BugBuddy popped up saying that evolution had crashed and gave me a backtrace. Here's the end of the E-D-S log, note the similar signature as previous postings to this bug: -------------------------------------------------------------------------- libecalbackendgroupwise-Message: ******** found libecalbackendgroupwise-Message: ******** found ***MEMORY-ERROR***: evolution-data-server[3890]: GSlice: assertion failed: sinfo->n_allocated > 0 -------------------------------------------------------------------------- Here's the end of the evolution log file: -------------------------------------------------------------------------- 200 SOAP-Debug: 0xa0c973d0 @ 1150411393 Pragma: no-cache Date: Thu, 15 Jun 2006 16:43:13 GMT Server: Windows NT GroupWise POA 7.0.1 Content-Length: 526 Content-Type: text/xml (evolution-2.6:3893): libecal-WARNING **: e-cal.c:1763: Unable to contact backend GThread-ERROR **: file gthread-posix.c: line 160 (): error 'Device or resource busy' during 'pthread_mutex_destroy ((pthread_mutex_t *) mutex)' aborting... <?xml version="1.0" encoding="UTF-8"?><SOAP-ENV:Envelope xmlns:xsi="http://www.w3.org/1999/XMLSchema-instance" xmlns:xsd="http://www.w3.org/1999/XMLSchema" xmlns:SOAP-ENC="http://schemas.xmlsoap.org/soap/encoding/" xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/"><SOAP-ENV:Header SOAP-ENV:encodingStyle=""/><SOAP-ENV:Body SOAP-ENV:encodingStyle="" xmlns:types="http://schemas.novell.com/2003/10/NCSP/types.xsd"><markReadResponse><status><code>0</code></status></markReadResponse></SOAP-ENV:Body></SOAP-ENV:Envelope> -------------------------------------------------------------------------- I'll attach the backtrace next.
Created attachment 67455 [details] Evolution crash after E-D-S crash when Tasks failed
Just tried again. And this time E-D-S crashed with the calendar backend first. No backtrace this time. "Your calendars will not be available until Evolution is restarted." "Your tasks will not be available until Evolution is restarted." Here's the end of the E-D-S logfile: ---------------------------------------------------------------------------- libecalbackendgroupwise-Message: ******** found libecalbackendgroupwise-Message: ******** found libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1457: Starting query ((and (occur-in-time-range? (make-time "20060528T060000Z") (make-time "20060806T060000Z")) #t)) libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1411: Getting object list ((and (occur-in-time-range? (make-time "20060528T060000Z") (make-time "20060806T060000Z")) #t)) ***MEMORY-ERROR***: evolution-data-server[3983]: GSlice: assertion failed: sinfo->n_allocated > 0 ----------------------------------------------------------------------------
And again. I forced evolution to shutdown. I restarted using the same commands as comment #17. Evolution came up in the calendar view. My Groupwise calendar was checked, but nothing was showing as appointments. I unclicked the Groupwise calendar and clicked it again. E-D-S crashed with the same messages as in comment #19. Evolution crashed too and I got another backtrace. Is this helpful? I don't mean to throw too much at you just to keep you busy... I would just really like to help find where this bug is. Here's the end of the E-D-S logfile: ---------------------------------------------------------------------------- libecalbackendgroupwise-Message: ******** found libecalbackendgroupwise-Message: ******** found libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1457: Starting query ((and (occur-in-time-range? (make-time "20060615T060000Z") (make-time "20060616T060000Z")) )) libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1411: Getting object list ((and (occur-in-time-range? (make-time "20060615T060000Z") (make-time "20060616T060000Z")) )) libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1457: Starting query ((and (occur-in-time-range? (make-time "20060612T060000Z") (make-time "20060617T060000Z")) )) libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1411: Getting object list ((and (occur-in-time-range? (make-time "20060612T060000Z") (make-time "20060617T060000Z")) )) libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1457: Starting query ((and (occur-in-time-range? (make-time "20060610T060000Z") (make-time "20060617T060000Z")) )) libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1411: Getting object list ((and (occur-in-time-range? (make-time "20060610T060000Z") (make-time "20060617T060000Z")) )) libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1457: Starting query ((and (occur-in-time-range? (make-time "20060528T060000Z") (make-time "20060709T060000Z")) )) libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1411: Getting object list ((and (occur-in-time-range? (make-time "20060528T060000Z") (make-time "20060709T060000Z")) )) libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1457: Starting query ((and (occur-in-time-range? (make-time "20060601T060000Z") (make-time "20060701T060000Z")) )) libecalbackendgroupwise-Message: e-cal-backend-groupwise.c:1411: Getting object list ((and (occur-in-time-range? (make-time "20060601T060000Z") (make-time "20060701T060000Z")) )) ***MEMORY-ERROR***: evolution-data-server[4102]: GSlice: assertion failed: sinfo->n_allocated > 0 ---------------------------------------------------------------------------- Here's the end of the evolution log file: -------------------------------------------------------------------------- 200 SOAP-Debug: 0x8547050 @ 1150412480 Pragma: no-cache Date: Thu, 15 Jun 2006 17:01:20 GMT Server: Windows NT GroupWise POA 7.0.1 Content-Length: 593 Content-Type: text/xml calendar-gui-Message: Check if default client matches (1104859343.1012.9@user.company.com 1133807875.8901.5@user.company.com) (evolution-2.6:4105): libecal-WARNING **: e-cal-view.c:391: Unable to start view -------------------------------------------------------------------------- I'll attach the new backtrace.
Created attachment 67456 [details] Evolution crash after E-D-S crash when Calendar failed
Created attachment 67501 [details] General E-D-S crash backtrace I found a Bug Buddy error notice when I first clicked in the evolution window today after it was left idle for 3+ hours. The error message said: The Application "/usr/libexec/evolution-data-server-1.6" has quit unexpectedly. This attachment is the backtrace obtained.
Created attachment 67519 [details] Another general E-D-S crash backtrace This E-D-S crash happened very soon after logging in and starting Gnome.
I have observed this crash happening frequently with evoltuion 2.7.3 with groupwise account configured. I have built evo on 22/06. Confirming it.
*** This bug has been marked as a duplicate of 347987 ***