GNOME Bugzilla – Bug 317956
Crash when starting evolution without or with expired kerberos 5 ticket
Last modified: 2005-10-10 19:16:36 UTC
Distribution: Ubuntu 5.10 (breezy) Package: Evolution-Data-Server Severity: critical Version: GNOME2.12.1 1.4.x Gnome-Distributor: Ubuntu Synopsis: Crash when starting evolution without or with expired kerberos 5 ticket Bugzilla-Product: Evolution-Data-Server Bugzilla-Component: Mailer Bugzilla-Version: 1.4.x BugBuddy-GnomeVersion: 2.0 (2.12.0) Description: I'm reading mail with imap and using gssapi for authentication. After upgrading to the latest version in Ubuntu Breezy today evolution crashed every time I try to start it without a kerberos 5 ticket or with an expired one. Unfortunately the BT is probably pretty useless. I will se what I can do to further debug it, unless it is an already known bug. Ubuntu version of evolution and evolution-data-server are: evolution 2.4.1-0ubuntu4 evolution-data-server 1.4.1-0ubuntu1 evolution-plugins 2.4.1-0ubuntu4 e-d-s is built against MIT Kerberos 1.3.6 if that does matter. /torkel Debugging Information: Backtrace was generated from '/usr/bin/evolution' (no debugging symbols found) Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1". (no debugging symbols found) `system-supplied DSO at 0xffffe000' has disappeared; keeping its symbols. (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) (no debugging symbols found) [Thread debugging using libthread_db enabled] [New Thread -1228694720 (LWP 2818)] [New Thread -1288393808 (LWP 2843)] [New Thread -1279870032 (LWP 2833)] [New Thread -1271075920 (LWP 2831)] [New Thread -1262683216 (LWP 2829)] [New Thread -1254253648 (LWP 2828)] [New Thread -1245860944 (LWP 2827)] [New Thread -1237468240 (LWP 2826)] (no debugging symbols found) 0xffffe410 in __kernel_vsyscall ()
+ Trace 63362
Thread 5 (Thread -1262683216 (LWP 2829))
------- Bug moved to this database by unknown@gnome.bugs 2005-10-04 19:39 UTC -------
When running with CAMEL_DEBUG=all I get the following before evolution crashes. adding hook target 'source' (evolution:4857): camel-WARNING **: camel_exception_get_id called with NULL parameter. received: * OK egg.hpc2n.umu.se Cyrus IMAP4 v2.0.16 server ready sending : A00000 CAPABILITY received: * CAPABILITY IMAP4 IMAP4rev1 ACL QUOTA LITERAL+ NAMESPACE UIDPLUS ID NO_ATOMIC_RENAME UNSELECT MULTIAPPEND SORT THREAD=ORDEREDSUBJECT THREAD=REFERENCES IDLE STARTTLS AUTH=PLAIN AUTH=GSSAPI AUTH=DIGEST-MD5 AUTH=CRAM-MD5 received: A00000 OK Completed sending : A00001 AUTHENTICATE GSSAPI received: + received: A00001 BAD Invalid base64 string sending : A00002 LOGOUT received: * BYE LOGOUT received (evolution:4857): Gdk-CRITICAL **: gdk_gc_set_foreground: assertion `GDK_IS_GC (gc)' failed received: * OK egg.hpc2n.umu.se Cyrus IMAP4 v2.0.16 server ready sending : A00000 CAPABILITY received: * CAPABILITY IMAP4 IMAP4rev1 ACL QUOTA LITERAL+ NAMESPACE UIDPLUS ID NO_ATOMIC_RENAME UNSELECT MULTIAPPEND SORT THREAD=ORDEREDSUBJECT THREAD=REFERENCES IDLE STARTTLS AUTH=PLAIN AUTH=GSSAPI AUTH=DIGEST-MD5 AUTH=CRAM-MD5 received: A00000 OK Completed sending : A00001 AUTHENTICATE GSSAPI received: + (evolution:4857): camel-WARNING **: camel_exception_get_id called with NULL parameter.
Created attachment 53078 [details] Somewhat better BT of the crash
Created attachment 53234 [details] [review] Fixes a crash if sasl_resp is NULL sasl_resp need to be checked if it is NULL and then bail out otherwise it will crash when running strlen on sasl_resp. Has strlen always segfaulted on a NULL string?
Thanks for the patch. Committed to HEAD.
the patch has a type-o in it, just FYI. Also, you should probably set an exception if sasl_resp is NULL
Ooops. That's what might happen when you test the patch with a tar-ball but does the diff against CVS and forget don't verify it. Thanks for pointing it out Jeff. Luckily enough the patch seems to have been applied with the typo fixed. Thanks Partha! What's wrong with using the fallback exception, "Bad authentication response from server."? What would be a better execption?
I guess the fallback exception is fine, I wasn't sure if one got set or not tho