Closed Bug 867099 Opened 13 years ago Closed 12 years ago

crash in ? (nsMsgDatabase::GetFolder)

Categories

(MailNews Core :: Backend, defect)

x86
Windows NT
defect
Not set
critical

Tracking

(thunderbird-esr17 affected)

RESOLVED WORKSFORME
Tracking Status
thunderbird-esr17 --- affected

People

(Reporter: wsmwk, Unassigned)

References

Details

(Keywords: crash, topcrash-thunderbird, Whiteboard: [regression:TB17.0.5,not][TB17.0.5,TB17.0.8])

Crash Data

TB17.0.5 regression. #7 crash Majority of stacks are of this form... bp-323735c7-8d69-4ac3-a7fa-971b62130404 ============================================================= 0 xul.dll nsMsgDatabase::GetFolder mailnews/base/src/nsMsgWindow.cpp:196 1 xul.dll nsASDOMWindowEnumerator::GetNext xpfe/appshell/src/nsAppShellWindowEnumerator.cpp:227 2 xul.dll MarkWindowList content/base/src/nsCCUncollectableMarker.cpp:253 3 xul.dll nsCCUncollectableMarker::Observe content/base/src/nsCCUncollectableMarker.cpp:314 4 xul.dll nsObserverList::NotifyObservers xpcom/ds/nsObserverList.cpp:99 5 xul.dll nsObserverService::NotifyObservers xpcom/ds/nsObserverService.cpp:149 6 xul.dll nsCycleCollector::ForgetSkippable xpcom/base/nsCycleCollector.cpp:2080 7 xul.dll nsCycleCollector_forgetSkippable xpcom/base/nsCycleCollector.cpp:3244 8 xul.dll FireForgetSkippable dom/base/nsJSEnvironment.cpp:3049 9 xul.dll CCTimerFired dom/base/nsJSEnvironment.cpp:3287 A few like bp-f844512c-04c1-471c-b9e4-1fa982130427 have nsMsgMailSession::GetTopmostMsgWindow on stack. Which are perhaps the same as https://crash-stats.mozilla.com/query/query?product=Thunderbird&version=ALL%3AALL&range_value=1&range_unit=weeks&date=04%2F30%2F2013+11%3A42%3A24&query_search=signature&query_type=contains&query=nsMsgMailSession%3A%3AGetTopmostMsgWindow&reason=&build_id=&process_type=any&hang_type=any&do_query=1 ?
#20 crash for esr 17.0.5 like bp-e9c1c8b5-4f52-40f3-8342-bbab22130427 some stack variation bp-cd773e81-f687-4502-a59c-a21bd2130430 No crashes found in development branches
Keywords: qawanted
Whiteboard: [regression:TB17.0.5 → [regression:TB17.0.5]
old version may be bug 694646... by the optimization by compiler, function symbol seems to be changed...
Yeah, GetFolder is bogus here, like Makoto says, this is probably a dupe of an older bug that's just morphed.
(In reply to Makoto Kato from comment #2) > old version may be bug 694646... Makoto, thanks. I agree. which raises the questions: - what was the signature in 17.0.4? (not terribly important - but nsMsgDatabase::GetFolder is not in 17.0.4 or 0.3) - what is the crash sig in Thunderbird beta? (nsMsgDatabase::GetFolder is not in beta, nor is nsXULWindow::GetDocShell) - but most importantly, is this bad cycle collector or bad thread behavior? or something else Unfortunately users I have contacted are not able to reproduce. like bp-b339b12d-1890-465f-b665-156462130414 (dr_schmidt) bp-f844512c-04c1-471c-b9e4-1fa982130427 (roy)
Depends on: 694646
Summary: crash in nsMsgDatabase::GetFolder → crash in ? (nsMsgDatabase::GetFolder)
Keywords: regression
Whiteboard: [regression:TB17.0.5] → [regression:TB17.0.5,not]
On pace to be a topcrash again, for TB17.0.8. signature disappeared in TB17.0.6 and TB17.0.7 17.0.2 nsXULWindow::GetDocShell(nsIDocShell**) 17.0.3 nsMimeBaseEmitter::GetOutputListener(nsIStreamListener**) bug 694646 17.0.4 nsMimeBaseEmitter::GetOutputListener(nsIStreamListener**) bug 694646 [17.0.5 nsMsgDatabase::GetFolder(nsIMsgFolder**) bug 867099] 17.0.6 nsMimeBaseEmitter::GetOutputListener(nsIStreamListener**) bug 694646 17.0.7 nsMsgDatabase::GetFolder(nsIMsgFolder**) bug 694646 17.0.8 nsMsgDatabase::GetFolder(nsIMsgFolder**) bug 867099
Whiteboard: [regression:TB17.0.5,not] → [regression:TB17.0.5,not][TB17.0.5,TB17.0.8]
Blocks: 867534
topcrash is being replaced by more precise keywords per https://bugzilla.mozilla.org/show_bug.cgi?id=927557#c3
Keywords: topcrash-win
No crashes newer than Thunderbird 17.0.8. Unclear to what extent bug 679626 helped without digging deep.
Status: NEW → RESOLVED
Closed: 12 years ago
Keywords: qawanted
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.