518674 - startup crash [@ nsLocalURI2Path(char const*, char const*, nsCString&)]

Status: VERIFIED FIXED

(MailNews Core :: Backend, defect)

Description

[Wayne Mery (:wsmwk)]

startup crash [@ nsLocalURI2Path(char const*, char const*, nsCString&)]

currently #7 3.0b4 crash, can't say if it will stay there.
in past month: only 1 crash for 3.0b3, so not even in top 100. but good many for 3.0b4pre and 3.0pre
no FF crashes


bp-850e184d-b0e2-4f7d-a4db-eab402090923
0	thunderbird.exe	nsLocalURI2Path	 mailnews/local/src/nsLocalUtils.cpp:184
1	thunderbird.exe	nsMailboxService::PrepareMessageUrl	mailnews/local/src/nsMailboxService.cpp:474
2	thunderbird.exe	nsMailboxService::FetchMessage	mailnews/local/src/nsMailboxService.cpp:242
3	thunderbird.exe	nsMailboxService::StreamMessage	mailnews/local/src/nsMailboxService.cpp:331
4	xpcom_core.dll	NS_InvokeByIndex_P	xpcom/reflect/xptcall/src/md/win32/xptcinvoke.cpp:101
5	xpcom_core.dll	xptiInterfaceEntry::GetIIDForParamNoAlloc	xpcom/reflect/xptinfo/src/xptiInterfaceInfo.cpp:479
6	thunderbird.exe	XPCWrappedNative::CallMethod	js/src/xpconnect/src/xpcwrappednative.cpp:2454
7	thunderbird.exe	XPC_WN_CallMethod	js/src/xpconnect/src/xpcwrappednativejsops.cpp:1590
8	js3250.dll	js_Invoke	js/src/jsinterp.cpp:1386
9	js3250.dll	js_Interpret	js/src/jsinterp.cpp:5179
10	js3250.dll	SendToGenerator	js/src/jsiter.cpp:857
11	js3250.dll	generator_op	js/src/jsiter.cpp:972
12	js3250.dll	generator_next	js/src/jsiter.cpp:987
13	js3250.dll	js_Interpret	js/src/jsinterp.cpp:5147
14	js3250.dll	SendToGenerator	js/src/jsiter.cpp:857
15	js3250.dll	generator_op	js/src/jsiter.cpp:972
16	js3250.dll	generator_send	js/src/jsiter.cpp:981
17	js3250.dll	js_Interpret	js/src/jsinterp.cpp:5147
18	js3250.dll	SendToGenerator	js/src/jsiter.cpp:857
19	js3250.dll	generator_op	js/src/jsiter.cpp:972
20	js3250.dll	generator_next	js/src/jsiter.cpp:987
21	js3250.dll	js_Interpret	js/src/jsinterp.cpp:5147
22	js3250.dll	js_Invoke	js/src/jsinterp.cpp:1394
23	thunderbird.exe	nsXPCWrappedJSClass::CallMethod	js/src/xpconnect/src/xpcwrappedjsclass.cpp:1697
24	thunderbird.exe	nsXPCWrappedJS::CallMethod	js/src/xpconnect/src/xpcwrappedjs.cpp:569

Comment 1

[Wayne Mery (:wsmwk)]

#1 for 3.0pre, so may as call this a topcrash and blocking

Comment 2

[David :Bienvenu]

gloda may be trying to stream messages w/ some sort of bad account setup. May need to bullet proof something.

Comment 3

[Mark Banner (:standard8)]

(In reply to comment #1)
> #1 for 3.0pre, so may as call this a topcrash and blocking

I disagree with the assessment of topcrash here.

There are 19 crashes in that list, all with the same build id/windows version and for at least 3 of them the dlls are exactly the same versions and list of.

Coupled with the fact that 18 of the crashes are within 12 minutes of each other, I really don't think this is a frequent crasher.

Whilst in the strictest sense it is a top crasher of 3.0pre over the last week, I think having the majority of crashes submitted by quite clearly the same user, negates that a bit.

Having said that, it appears to be relatively high on a few of the various lists I've looked at, so we should possibly look at fixing this anyway.

Comment 4

[David :Bienvenu]

If this really is triggered by gloda trying to stream the same message, this might make it impossible to use TB for these set of users...

Comment 5

[Wayne Mery (:wsmwk)]

(In reply to comment #3)
> (In reply to comment #1)
> > #1 for 3.0pre, so may as call this a topcrash and blocking
> 
> I disagree with the assessment of topcrash here.
> 
> There are 19 crashes in that list, all with the same build id/windows version
> and for at least 3 of them the dlls are exactly the same versions and list of.
> 
> Coupled with the fact that 18 of the crashes are within 12 minutes of each
> other, I really don't think this is a frequent crasher.

yeah. quite probably one darned persistent person :)
when I saw that many crashes I didn't consider it could be one bloke.
too bad we don't have metrics on unique crash submitters yet.

Comment 6

[David :Bienvenu]

Attachment: proposed fix

the sole caller seems to check the result so this should fix the crash.

Comment 7

[David :Bienvenu]

fix checked in.

Comment 8

[Wayne Mery (:wsmwk)]

v.fixed
no crashes except 3.0b4