inconsistency within storage.sdb causes Lightning to segfault (in



Lightning Only
11 years ago
11 years ago


(Reporter: Wolfgang Sourdeau, Unassigned)




(1 attachment)



11 years ago
User-Agent:       Mozilla/5.0 (X11; U; Linux i686; fr; rv: Gecko/20061205 Iceweasel/ (Debian-
Build Identifier: cvs:2007-02-01 1_8_BRANCH

all is in the summary, the erroneous storage.sdb will be attached to this bugreport. Even if the file got somehow "corrupted", Lightning should not choke.

Reproducible: Always

Steps to Reproduce:

the generated backtrace is as follows:

(gdb) r
Starting program: /usr/lib/icedove/icedove-bin
[Thread debugging using libthread_db enabled]
[New Thread -1220921664 (LWP 32043)]
[New Thread -1223775312 (LWP 32050)]
[New Thread -1241613392 (LWP 32053)]
[New Thread -1257464912 (LWP 32056)]
DOUBLE-CLICK: 400 --> -1 THRESHOLD: 8 --> -1 [New Thread -1266234448 (LWP 32057)]
[New Thread -1274627152 (LWP 32058)]
[New Thread -1284957264 (LWP 32059)]
Thank you for choosing Reloadchromezilla
Guessing system timezone:
TZoffset1: -0400
TZoffset2: -0500
TZname1: EDT
TZname2: EST
[New Thread -1297802320 (LWP 32062)]
[New Thread -1306195024 (LWP 32063)]
[Thread -1306195024 (LWP 32063) exited]
[New Thread -1314587728 (LWP 32064)]
[Thread -1314587728 (LWP 32064) exited]
*** Calendar schema version is: 6
[New Thread -1314587728 (LWP 32065)]

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread -1220921664 (LWP 32043)]
0xb2ac0cd3 in calDateTime::GetIcalTZ (this=0x8cad198, tzid=@0x8cad338, tzp=0xbfaf6508) at calDateTime.cpp:683
683         nsresult rv = ics->GetTimezone(tzid, getter_AddRefs(tz));
(gdb) bt
#0  0xb2ac0cd3 in calDateTime::GetIcalTZ (this=0x8cad198, tzid=@0x8cad338, tzp=0xbfaf6508) at calDateTime.cpp:683
#1  0xb2ac2406 in calDateTime::GetInTimezone (this=0x8cad198, aTimezone=@0x8cad338, aResult=0xbfaf66a0)
    at calDateTime.cpp:361
#2  0xb7e30379 in XPTC_InvokeByIndex () at xptcinvoke_gcc_x86_unix.cpp:50
#3  0xb6806b59 in XPCWrappedNative::CallMethod (ccx=@0xbfaf67bc, mode=XPCWrappedNative::CALL_METHOD)
    at xpcwrappednative.cpp:2156
#4  0xb680d47a in XPC_WN_CallMethod (cx=0x8346b30, obj=0x8ca63f0, argc=1, argv=0x8a9aecc, vp=0xbfaf68dc)
    at xpcwrappednativejsops.cpp:1451
#5  0xb7e9ccae in js_Invoke (cx=0x8346b30, argc=1, flags=0) at jsinterp.c:1187
#6  0xb7e94d58 in js_Interpret (cx=0x8346b30, pc=0x8548906 ":", result=0xbfaf6c38) at jsinterp.c:3586
#7  0xb7e9cdd6 in js_Invoke (cx=0x8346b30, argc=1, flags=2) at jsinterp.c:1207
#8  0xb6804b65 in nsXPCWrappedJSClass::CallMethod (this=0x84cafe8, wrapper=0x8547038, methodIndex=3, info=0x81ae1e0,
    nativeParams=0xbfaf6f0c) at xpcwrappedjsclass.cpp:1410
#9  0xb67feec0 in nsXPCWrappedJS::CallMethod (this=0x8547038, methodIndex=3, info=0x81ae1e0, params=0xbfaf6f0c)
    at xpcwrappedjs.cpp:461
#10 0xb7e30f41 in PrepareAndDispatch (methodIndex=<value optimized out>, self=0x8547038, args=<value optimized out>)
    at xptcstubs_gcc_x86_unix.cpp:100
#11 0xb639f4a2 in nsEventListenerManager::HandleEventSubType (this=0x83e7618, aListenerStruct=0x8515bc0,
    aDOMEvent=0x8ae2db8, aCurrentTarget=0x834692c, aSubType=1, aPhaseFlags=7) at nsEventListenerManager.cpp:1687
#12 0xb63a07c4 in nsEventListenerManager::HandleEvent (this=0x83e7618, aPresContext=0x8390d70, aEvent=0xbfaf7228,
    aDOMEvent=0xbfaf71d8, aCurrentTarget=0x834692c, aFlags=7, aEventStatus=0xbfaf7280)
    at nsEventListenerManager.cpp:1788
#13 0xb64a9e46 in nsGlobalWindow::HandleDOMEvent (this=0x835af50, aPresContext=0x8390d70, aEvent=0xbfaf7228,
    aDOMEvent=0xbfaf71d8, aFlags=7, aEventStatus=0xbfaf7280) at nsGlobalWindow.cpp:1598
#14 0xb61b9a1c in DocumentViewerImpl::LoadComplete (this=0x821c028, aStatus=0) at nsDocumentViewer.cpp:1012
#15 0xb5758a62 in nsDocShell::EndPageLoad (this=0x83461b0, aProgress=0x83461c4, aChannel=0x8371280, aStatus=0)
    at nsDocShell.cpp:4802
#16 0xb575f6ae in nsWebShell::EndPageLoad (this=0x83461b0, aProgress=0x83461c4, channel=0x8371280, aStatus=0)
    at nsWebShell.cpp:659
#17 0xb5758f6f in nsDocShell::OnStateChange (this=0x83461b0, aProgress=0x83461c4, aRequest=0x8371280,
    aStateFlags=131088, aStatus=0) at nsDocShell.cpp:4728
#18 0xb576886d in nsDocLoader::FireOnStateChange (this=0x83461b0, aProgress=0x83461c4, aRequest=0x8371280,
    aStateFlags=131088, aStatus=0) at nsDocLoader.cpp:1210
#19 0xb5768948 in nsDocLoader::doStopDocumentLoad (this=0x83461b0, request=0x8371280, aStatus=0) at nsDocLoader.cpp:833
#20 0xb5768e48 in nsDocLoader::DocLoaderIsEmpty (this=0x83461b0) at nsDocLoader.cpp:739
#21 0xb5769386 in nsDocLoader::OnStopRequest (this=0x83461b0, aRequest=0x83d13f0, aCtxt=0x0, aStatus=0)
    at nsDocLoader.cpp:662
#22 0xb7146230 in nsLoadGroup::RemoveRequest (this=0x8346570, request=0x83d13f0, ctxt=0x0, aStatus=0)
    at nsLoadGroup.cpp:732
#23 0xb636277e in nsDocument::UnblockOnload (this=0x83d1738) at nsDocument.cpp:5015
#24 0xb6327c3c in HandleASyncMenuGeneration (aEvent=0x8b15cd8) at nsMenuFrame.cpp:2059
#25 0xb7e170d7 in PL_HandleEvent (self=0x8b15cd8) at plevent.c:688
#26 0xb7e173e9 in PL_ProcessPendingEvents (self=0x80c6f90) at plevent.c:623
#27 0xb7e1926e in nsEventQueueImpl::ProcessPendingEvents (this=0x80ef4a8) at nsEventQueue.cpp:417
#28 0xb67a4f75 in event_processor_callback (source=0x837ff60, condition=G_IO_IN, data=0xbfaf648c) at nsAppShell.cpp:67
#29 0xb774006f in ?? () from /usr/lib/
#30 0x0837ff60 in ?? ()
#31 0x00000001 in ?? ()
#32 0x080ef4a8 in ?? ()
#33 0xb777c1e4 in ?? () from /usr/lib/
#34 0xb7d51400 in ?? () from /lib/tls/i686/cmov/
#35 0x0837ffa8 in ?? ()
#36 0xbfaf7978 in ?? ()
#37 0xb7716b21 in g_main_context_dispatch () from /usr/lib/
Backtrace stopped: frame did not save the PC

Comment 1

11 years ago
Created attachment 253645 [details]
the faulty storage.sdb

Maybe the problem is related to the attached db along with another preference setting, if so, please let me know.
Keywords: qawanted
[2007-02-01 20:29:58] <Wolfgang-> my storage.sdb was somehow corrupted, I will post it along my bugreport
[2007-02-01 20:41:14] <Wolfgang->
[2007-02-01 20:45:02] <lilmatt> storage.dbs is it
[2007-02-01 20:45:04] <lilmatt> sdb
[2007-02-01 20:45:34] <Wolfgang-> ok
[2007-02-01 20:58:48] <ssitter> Wolfgang-: if i copy and paste the storage.sdb into my profile i don't see a crash
[2007-02-01 20:59:33] <ssitter> could the event that causes the crash be stored in one of the remote calendars you use?
[2007-02-01 21:00:00] <Wolfgang-> it's not a remote event because the app doesn't even ask me for a password before crashing.
[2007-02-01 21:00:29] <Wolfgang-> I could send you my prefs?
[2007-02-01 21:03:58] <ssitter> i see only one event there, is that correct?
[2007-02-01 21:04:38] <Wolfgang-> yes, it's possible
[2007-02-01 21:05:02] <Wolfgang-> mmmm.... could it be the problem occurs when an event has no timezone specified (using the "Z" suffix)?
[2007-02-01 21:05:41] <ssitter> the event in storage.sdb seems to use '/'
[2007-02-01 21:05:49] <Wolfgang-> ok
[2007-02-01 21:07:54] <Wolfgang-> you think it's related to that event?
[2007-02-01 21:08:15] <Wolfgang-> I am not sure it even reaches that point of initialization before crashing actually...
[2007-02-01 21:08:51] <ssitter> i don't think so, the .sdb seems fine here using sb 0.4a1 on windows
[2007-02-01 21:09:07] <Wolfgang-> nightly build?
[2007-02-01 21:09:13] <ssitter> yes
[2007-02-01 21:09:17] <Wolfgang-> ok
[2007-02-01 21:10:53] <ssitter> maybe you can try a new profile and subscribe to the caldav calendars again to see if one of those causes the problem or did you already tried?
[2007-02-01 21:10:56] <ssitter> ( i don't read the discussion from beginning)
[2007-02-01 21:11:36] <Wolfgang-> no, I killed my storage.sdb altogether and it seems to be working fine... (at least in its current uninitialized state)
I copied the storage.sdb into my profile and see no crash -> Works for me using Lightning/0.5pre (2007032303) with Mozilla/5.0 (X11; U; Linux i686; en-US; rv: Gecko/20070323 Thunderbird/2.0pre ID:2007032303
Keywords: qawanted
Whiteboard: [qa discussion needed]

Comment 4

11 years ago
Wolfgang, do you still see this behavior on a current nightly build?  If not, we will go ahead and closet this defect.
Whiteboard: [qa discussion needed]
Marking WFM per comment#3 and no answer from reporter. Feel free to reopen this bug if the problem recurs.
Last Resolved: 11 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.