Closed Bug 411147 Opened 18 years ago Closed 6 years ago

crash [@ nsQueryInterface::operator()(nsID const&, void**)] (windows only)

Categories

(Thunderbird :: General, defect)

x86
Windows
defect
Not set
critical

Tracking

(Not tracked)

RESOLVED INCOMPLETE

People

(Reporter: wsmwk, Unassigned)

References

Details

(Keywords: crash, Whiteboard: [need skiplist for nsQueryInterface::operator()])

Crash Data

Thunderbird crash, windows only per crash-stats. Maybe a windows variation of bug 411145, which is "nsQueryInterface::operator()(nsID const&, void**) const" and on crash-stats is Mac and linux only. began around 2007092900. frequency seems to have escalated as of 2007121500. I've had several myself in the past couple weeks. some call out the frame nsQueryInterface::operator()(nsID const&, void**), but some start with an address (with rest of stack being the same) today's example is bp-597f8cca-bd2f-11dc-be9c-001a4bd43e5c. reproducible: no steps prior to this specific crash: 1. advanced search from a previous search (i.e. already open with one search term) 2. click "+" to add a second search term 3. click in new input box, I didn't type anything 4. back tab (not click) once or twice 5. crash Signature nsQueryInterface::operator()(nsID const&, void**) UUID 597f8cca-bd2f-11dc-be9c-001a4bd43e5c Time 2008-01-07 06:45:25-08:00 Build ID 2007123103 OS Windows NT OS Version 5.1.2600 Service Pack 2 CPU x86 CPU Info GenuineIntel family 15 model 2 stepping 9 Crash Reason EXCEPTION_ACCESS_VIOLATION Crash Address 0x0 0 nsQueryInterface::operator()(nsID const&, void**) nsCOMPtr.cpp:47 1 thunderbird.exe@0x795ae3 2 nsCOMPtr<nsIContent>::nsCOMPtr<nsIContent>(nsQueryInterface) nsCOMPtr.h:645 3 nsXULDocument::AttributeChanged(nsIDocument*, nsIContent*, int, nsIAtom*, int, unsigned int) mozilla/content/xul/document/src/nsXULDocument.cpp:956 4 nsNodeUtils::AttributeChanged(nsIContent*, int, nsIAtom*, int, unsigned int) mozilla/content/base/src/nsNodeUtils.cpp:109 5 nsGenericElement::SetAttrAndNotify(int, nsIAtom*, nsIAtom*, nsAString_internal const&, nsAttrValue&, int, int, int) mozilla/content/base/src/nsGenericElement.cpp:3710 6 nsGenericElement::SetAttr(int, nsIAtom*, nsIAtom*, nsAString_internal const&, int) mozilla/content/base/src/nsGenericElement.cpp:3641 7 nsGenericElement::SetAttribute(nsAString_internal const&, nsAString_internal const&) mozilla/content/base/src/nsGenericElement.cpp:1533 8 NS_InvokeByIndex_P mozilla/xpcom/reflect/xptcall/src/md/win32/xptcinvoke.cpp:101 9 AutoJSSuspendRequest::SuspendRequest() mozilla/js/src/xpconnect/src/xpcprivate.h:3407 10 nsSelectMoveScrollCommand::DoCommandBrowseWithCaretOn(char const*, nsISelectionController*, nsIEventStateManager*) mozilla/dom/src/base/nsGlobalWindowCommands.cpp:291
top 25 crash for v3.1.2 however, stacks vary. for example bp-3fc30d45-468f-4353-baa1-ac2772100812 bp-4775b96d-d579-4389-9363-f398d2100809 bp-81791686-c221-4557-a12f-620872100812 bp-d9bdde1e-468f-4815-b970-c07952100811 so we probably need a skiplist for nsQueryInterface::operator() so we don't need to rely on manual efforts to track unique stacks.
Keywords: qawanted, topcrash
Whiteboard: [need skiplist for nsQueryInterface::operator()]
Crash Signature: [@ nsQueryInterface::operator()(nsID const&, void**)]
still topcrash, with the addition of _purecall | nsQueryInterface::operator()(nsID const&, void**)
Crash Signature: [@ nsQueryInterface::operator()(nsID const&, void**)] → [@ nsQueryInterface::operator()(nsID const&, void**)] [@ _purecall | nsQueryInterface::operator()(nsID const&, void**)]
It's #63 top crasher in TB 17.0.2 so not a top crasher according to https://wiki.mozilla.org/CrashKill/Topcrash#Top_crash_identification_criteria
Keywords: topcrash
#96 crash for TB31, so still exists who users crashing with address 0x5a5a5a5a bp-e7e847eb-ab66-47e4-b290-e9cc82140826 dcarter pb-8abe4327-7a74-431e-bc8a-fba462140829 jean.bidault 0 xul.dll nsQueryInterface::operator()(nsID const&, void**) xpcom/glue/nsCOMPtr.cpp 1 xul.dll nsCOMPtr_base::assign_from_qi(nsQueryInterface, nsID const&) xpcom/glue/nsCOMPtr.cpp 2 xul.dll nsXULElement::GetBuilder() content/xul/content/src/nsXULElement.cpp 3 xul.dll nsXULElement::GetDatabase() content/xul/content/src/nsXULElement.cpp 4 xul.dll nsXULElement::GetDatabase(nsIRDFCompositeDataSource**) content/xul/content/src/nsXULElement.cpp 5 xul.dll mozilla::dom::XULDocument::CheckTemplateBuilderHookup(nsIContent*, bool*) content/xul/document/src/XULDocument.cpp
Flags: needinfo?(gibran.santos)
Flags: needinfo?(gibran.santos)
Crash Signature: [@ nsQueryInterface::operator()(nsID const&, void**)] [@ _purecall | nsQueryInterface::operator()(nsID const&, void**)] → [@ nsQueryInterface::operator()(nsID const&, void**)] [@ _purecall | nsQueryInterface::operator()(nsID const&, void**)] [@ nsQueryInterface::operator()] [@ _purecall | nsQueryInterface::operator()]
#56 crash for 45.7.0. I don't think these stacks/this bug are actionable on their own. However, most users I checked have other crash signatures involving "shutdown xxx", like bug 1307963.
OS: Windows XP → Windows
See Also: → 1307963
Eyballing last 7 days the crash rate of 52 vs 60 look rougly the same. None of the ones I checked have enigmail. Also, crash rate of last 6 months across all versions remains the same, peaks at ~64 crashes per day https://crash-stats.mozilla.com/signature/?product=Thunderbird&signature=nsQueryInterface%3A%3Aoperator%28%29&date=%3E%3D2018-05-24T15%3A37%3A44.000Z&date=%3C2018-11-24T14%3A37%3A44.000Z#graphs (In reply to Wayne Mery (:wsmwk) from comment #6) > #56 crash for 45.7.0. I don't think these stacks/this bug are actionable on > their own. However, most users I checked have other crash signatures > involving "shutdown xxx", like bug 1307963. That was recently tied to bug 1410286 - enigmail.
Crash Signature: [@ nsQueryInterface::operator()(nsID const&, void**)] [@ _purecall | nsQueryInterface::operator()(nsID const&, void**)] [@ nsQueryInterface::operator()] [@ _purecall | nsQueryInterface::operator()] → [@ nsQueryInterface::operator()] [@ _purecall | nsQueryInterface::operator()]
Keywords: qawanted

nsQueryInterface::operator() doesn't exist for version 68 and we've never had information that describes any context or how to reproduce this. It's either gone or signature morphed to something else

Status: NEW → RESOLVED
Closed: 6 years ago
Resolution: --- → INCOMPLETE
You need to log in before you can comment on or make changes to this bug.