|Summary:||Kicker segfaults in qt code, trying to close panel, when exiting kde.|
|Product:||[Fedora] Fedora||Reporter:||Mickey Stein <yekkim>|
|Component:||qt||Assignee:||Ngo Than <than>|
|Status:||CLOSED UPSTREAM||QA Contact:||Ben Levenson <benl>|
|Fixed In Version:||Doc Type:||Bug Fix|
|Doc Text:||Story Points:||---|
|Last Closed:||2005-06-18 06:03:27 UTC||Type:||---|
|oVirt Team:||---||RHEL 7.3 requirements from Atomic Host:|
|Cloudforms Team:||---||Target Upstream Version:|
Description Mickey Stein 2005-06-17 00:00:32 UTC
From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.8) Gecko/20050524 Fedora/1.0.4-4 Firefox/1.0.4 Description of problem: Since kde 3.4* hit rawhide, I've had some unusual behavour when exiting kde to the shell. (I stay in init 3 normally and start kde via 'startx'). Until the last update to qt (noted version above), I would intermittently see the panel appear for about 3-4 seconds before the usual exit to the shell. If this happened I'd see a message like this: -- DCOPClient::attachInternal. Attach failed Could not open netw ork socket DCOPClient::attachInternal. Attach failed Could not open network socket kicker: Fatal IO error: client killed ICE default IO error handler doing an exit(), pid = 26026, errno = 32 -- But most of the time it worked without the error and normally exited. I tried to move ~/.kde(settings) away but the problem persisted with a clean slate. After the recent (yesterday or the day before (13-16 Jun) qt update on rawhide, I now get a repeatable segfault kicker crash 'dialog' box which asks me if I want to save it before exiting. This is what I've gotten from it thus far: "no debugging symbols found) [KCrash handler] #3 0x081b12a7 in ?? () #4 0xb70f813e in QPtrList<QPopupMenu>::deleteItem () from /usr/lib/qt-3.3/lib/libqt-mt.so.3 #5 0xb728b51d in QGList::clear () from /usr/lib/qt-3.3/lib/libqt-mt.so.3 #6 0xb65401c9 in PanelServiceMenu::~PanelServiceMenu$base () from /usr/lib/libkdeinit_kicker.so #7 0xb653bca7 in PanelKMenu::~PanelKMenu$delete () from /usr/lib/libkdeinit_kicker.so #8 0xb6513f3d in MenuManager::~MenuManager$delete () from /usr/lib/libkdeinit_kicker.so #9 0xb6fa092c in QObject::~QObject$base () from /usr/lib/qt-3.3/lib/libqt-mt.so.3 #10 0xb6f409f5 in QApplication::~QApplication$base () from /usr/lib/qt-3.3/lib/libqt-mt.so.3 #11 0xb76b9dfe in KApplication::~KApplication$base () from /usr/lib/libkdecore.so.4 #12 0xb775c5fd in KUniqueApplication::~KUniqueApplication$base () from /usr/lib/libkdecore.so.4 #13 0xb64f6928 in Kicker::~Kicker () from /usr/lib/libkdeinit_kicker.so #14 0xb64f63f1 in kdemain () from /usr/lib/libkdeinit_kicker.so #15 0xb65986e0 in kdeinitmain () from /usr/lib/kde3/kicker.so #16 0x0804de4d in ?? () #17 0x0804e4c5 in ?? () #18 0x0804eaad in ?? () #19 0x0804fb29 in ?? () #20 0xb69a2dfb in __libc_start_main () from /lib/libc.so.6 #21 0x0804b911 in ?? () " I know that isn't very good information but I haven't been able to dig up the qt-debuginfo-3.3.4-15 file (oops ! Never mind, I just found it in .. 'debug' folder on the fedora site ;) ) I'll see if I can produce a better crash log and report back with that. Mick Version-Release number of selected component (if applicable): kdebase-3.4.1-0 & qt-3.3.4-15 How reproducible: Always Steps to Reproduce: 1. Start kde from init level 3 via 'startx'. 2. Do anything or nothing in kde 3. Exit kde. 4. Look for segfault info box. Actual Results: I get a crashdump in a kde box. Expected Results: I expect it to exit to a fullscreen shell as normal. It does this after I close the crash info box. Additional info: I'm using 'todays' kernel (2.6.12-rc6-git8) and all other software is up-to-date from rawhide.
Comment 1 Mickey Stein 2005-06-18 06:03:27 UTC
This was probably a bad choice of where to file the bug. I moved it upstream to bugs.kde.org and found a couple of identical already filed reports. 104152 is closest, and this is 'supposed' to be fixed in the next release of kdebase (3.4.2). I'll close this out here.
Comment 2 Ngo Than 2005-06-21 10:33:19 UTC
*** Bug 161198 has been marked as a duplicate of this bug. ***