GNOME Bugzilla – Bug 214218
occassional hangs on startup
Last modified: 2013-09-10 14:02:36 UTC
Package: Evolution Priority: Blocker Version: 0.16.99 Synopsis: occassional hangs on startup Bugzilla-Product: Evolution Bugzilla-Component: Miscellaneous Description: Occassionally (50% of the time) evolution hangs somewhere in startup. killev and trying again seems to get past this. Latest snapshot (cvs.2001.10.31.22.56) Unknown reporter: georgn@somanetworks.com, changed to bugbuddy-import@ximian.com.
*** bug 214219 has been marked as a duplicate of this bug. ***
Confirming; there are still reports of this in IRC and evolution@.
*** bug 214241 has been marked as a duplicate of this bug. ***
Forgot to cc ettore.
Created attachment 40483 [details] Traffic trace IMAP exchange <-> evolution
Created attachment 40484 [details] bt after ctrl-c in evolution-mail
The IMAP stuff is irrelevant here: it's hanging in ibex trying to unselect /local/Inbox
I got the following from the console: evolution-shell-WARNING **: Error setting owner on component OAFIID:GNOME_Evolution_Mail_ShellComponent -- CORBA-Fehler evolution-shell-WARNING **: Error changing interactive status of component OAFIID:GNOME_Evolution_Mail_ShellComponent to TRUE -- IDL:CORBA/COMM_FAILURE:1.0 evolution-shell-WARNING **: Error changing interactive status of component OAFIID:GNOME_Evolution_Mail_ShellComponent to FALSE -- IDL:CORBA/COMM_FAILURE:1.0 and after a gdb-run: gdb evolution-mail GNU gdb 5.0 Copyright 2000 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i386-suse-linux"... (gdb) r Starting program: /usr/bin/evolution-mail [New Thread 1024 (LWP 6673)] e-utils-ERROR **: Very serious error, cannot activate private config database 'Unknown CORBA exception id: 'IDL:OAF/GeneralError:1.0'' aborting... Program received signal SIGABRT, Aborted.
+ Trace 12805
Thread 1024 (LWP 6673)
hope that helps... Thomas
The last is another issue. The previous looks like the deadlock i fixed yesterday? Is that snapshot after 4pm yesterday?
*** bug 214256 has been marked as a duplicate of this bug. ***
*** bug 214232 has been marked as a duplicate of this bug. ***
So i find that by looking at the oh-so-well-documented g_* functions that g_mutex_trylock returns inverse error status to pthread_mutex_trylock. So i dont use g_mutex's anymore. Patch attached.
Created attachment 40491 [details] [review] Patch to remove g_mutex snot
Created attachment 40492 [details] [review] linux's destroy is a nop, forgot others might not be
FWIW, I'd be happier if we don't mark these fixed until the patch is applied. I don't want another 'I fixed it' 'oops, did I forget to apply that patch' situation like we had with other developers who shall go unnamed *cough*shell*cough*
Well its applied now.