91996 - Downloading the file second time crashes.

Status: VERIFIED FIXED

(SeaMonkey :: General, defect)

Description

[Suresh]

steps:
1. Download a file. (I downloaded today's trunk build to c:\temp.)
2. Download the same file again to the same location. "Do you wanna replace..."
dialog comes up and click ok. Boom..crahes.

Build: Todays commercial BRANCH build on Win 2K. consistently able to reproduce.

PS: Please change this bug to app. component. Thanks!!


Stack trace:
Incident ID 33249709
Stack Signature ntdll.dll + 0x4b9b1 (0x77fcb9b1) 7ae93fef
Bug ID
Trigger Time 2001-07-23 15:45:36
User Comments Saving a file second time crashes.......
Build ID 2001072306
Product ID Netscape6.10
Platform ID Win32
Stack Trace
ntdll.dll + 0x4b9b1 (0x77fcb9b1)
ntdll.dll + 0x4b733 (0x77fcb733)
MSVCRT.DLL + 0x1d92 (0x78001d92)
MSVCRT.DLL + 0xb2de (0x7800b2de)
DOMGCCallback [d:\builds\seamonkey\mozilla\dom\src\base\nsJSEnvironment.cpp,
line 1481]
js_GC [d:\builds\seamonkey\mozilla\js\src\jsgc.c, line 1305]
js_ForceGC [d:\builds\seamonkey\mozilla\js\src\jsgc.c, line 944]
js_DestroyContext [d:\builds\seamonkey\mozilla\js\src\jscntxt.c, line 228]
JS_DestroyContext [d:\builds\seamonkey\mozilla\js\src\jsapi.c, line 884]
nsJSContext::~nsJSContext
[d:\builds\seamonkey\mozilla\dom\src\base\nsJSEnvironment.cpp, line 390]
nsJSContext::`scalar deleting destructor'
nsJSContext::Release
[d:\builds\seamonkey\mozilla\dom\src\base\nsJSEnvironment.cpp, line 397]
nsCOMPtr_base::~nsCOMPtr_base
[d:\builds\seamonkey\mozilla\xpcom\base\nsCOMPtr.cpp, line 50]
GlobalWindowImpl::HandleDOMEvent
[d:\builds\seamonkey\mozilla\dom\src\base\nsGlobalWindow.cpp, line 648]
DocumentViewerImpl::LoadComplete
[d:\builds\seamonkey\mozilla\content\base\src\nsDocumentViewer.cpp, line 1126]
nsDocShell::EndPageLoad
[d:\builds\seamonkey\mozilla\docshell\base\nsDocShell.cpp, line 3731]
nsWebShell::EndPageLoad
[d:\builds\seamonkey\mozilla\docshell\base\nsWebShell.cpp, line 912]
nsDocShell::OnStateChange
[d:\builds\seamonkey\mozilla\docshell\base\nsDocShell.cpp, line 3652]
nsDocLoaderImpl::FireOnStateChange
[d:\builds\seamonkey\mozilla\uriloader\base\nsDocLoader.cpp, line 1095]
nsDocLoaderImpl::doStopDocumentLoad
[d:\builds\seamonkey\mozilla\uriloader\base\nsDocLoader.cpp, line 734]
nsDocLoaderImpl::DocLoaderIsEmpty
[d:\builds\seamonkey\mozilla\uriloader\base\nsDocLoader.cpp, line 632]
nsDocLoaderImpl::OnStopRequest
[d:\builds\seamonkey\mozilla\uriloader\base\nsDocLoader.cpp, line 563]
nsLoadGroup::RemoveRequest
[d:\builds\seamonkey\mozilla\netwerk\base\src\nsLoadGroup.cpp, line 517]
nsCachedChromeChannel::HandleStopLoadEvent
[d:\builds\seamonkey\mozilla\rdf\chrome\src\nsChromeProtocolHandler.cpp, line 446]
PL_HandleEvent [d:\builds\seamonkey\mozilla\xpcom\threads\plevent.c, line 591]
PL_ProcessPendingEvents [d:\builds\seamonkey\mozilla\xpcom\threads\plevent.c,
line 524]
_md_EventReceiverProc [d:\builds\seamonkey\mozilla\xpcom\threads\plevent.c, line
1072]
nsAppShellService::Run
[d:\builds\seamonkey\mozilla\xpfe\appshell\src\nsAppShellService.cpp, line 426]
netscp6.exe + 0x16f0 (0x004016f0)
netscp6.exe + 0x11b8 (0x004011b8)
netscp6.exe + 0x3366 (0x00403366)
KERNEL32.DLL + 0x17d08 (0x77e97d08)

Comment 1

[Suresh]

nominating

Comment 2

[Suresh]

Adding Vishy to the cc list, so that he can reassign this bug to the right owner
and component. Thanks!!

Comment 3

[Viswanath Ramachandran]

Blake - can you try this out? 
Suresh - did you see this with a debug or release build?

also adding jst due to DOM on the stack and Lisa. 

Comment 4

[Suresh]

Vishy - I saw this using release build. (Today's BRANCH Commercial build). I'm
updating my debug build...will update soon. But definetely it is 100% reproducible. 

Comment 5

[Blake Ross]

I can't reproduce in today's branch comm nightly on win2k.  I clicked on the
Mozilla 0.9.2 Windows link on mozilla.org and saved it in D:\. When it finished
downloading, I clicked it again, chose to save it in D:\, pressed Yes in the
confirmation, and it downloaded fine.

Comment 6

[Suresh]

hmm....here are the various senarios I have tried so far. *ALL RESULTED IN A
CRASH*.  I see 2-4 different stack traces. 

1. I tried to download M0.9.2 from mozilla.org. I did it twice and didn't see a
crash. Then I went to ftp://sweetlou....and clicked on today's branch link
(2001-07-23-06-0.9.2). Crashes. 

2. Went to my publich directory where I have which.exe file
(jazz/users/suresh/publish). I downloaded couple of times and went to mail,
switched to another folder...Crashes. If you don't see a crash try switching to
another folder and read some msgs.

Comment 7

[lchiang]

stephend - can you try this bug out to see if you can reproduce it?

Comment 8

[[e-mail] (gone - use [e-mail] instead)]

Hmmm..interesting.  I'll try to reproduce this again, but wanted to get this
stack and info to this report before Moz took a dive again! :-(  I couldn't
reproduce using Suresh's scenario (the 1st one).  What I did, and crashed with,
was downloaded a 34 meg file (Paint Shop Pro 7) from tucows.com, then while that
file was downloading, I attempted to download the same file again.  The XUL
window (saving file) appeared, but was blank, until the 1st file was finished
downloading, and moving from the salted temp directory onto my Desktop, where I
had specified it to save to.  When the 2nd window tried to populate itself with
the file's info, I crashed.  I'll post a stack.

Comment 9

[[e-mail] (gone - use [e-mail] instead)]

Attachment: My stack.

Comment 10

[[e-mail] (gone - use [e-mail] instead)]

Rats.  Doing my same exact steps, I couldn't reproduce this crash again.  I'll
keep poking at it though.  (this was with a restart)

Comment 11

[[e-mail] (gone - use [e-mail] instead)]

more info yet: This time, I wasn't downloading anything, but clicked on my
inbox, and just read a bugzilla mail.  Crashed, same stack as suresh and my other.

Comment 12

[lchiang]

vishy or nisheeth - any ideas of a possible owner?   The crash seems to be in js
area.

Comment 13

[[e-mail] (gone - use [e-mail] instead)]

Could it be jband/dbradley's fix to mozilla/ dom/ src/ base/ nsJSEnvironment.cpp
(1.146)?

Comment 14

[Suresh]

fyi: I'm able to duplicate this crash using BRANCH Debug build on Win 2K. 

Comment 15

[Suresh]

I backed out changes to bug 87389 and still able to duplicate this problem. Here
is the stack trace I got using debug build.

nsCOMPtr<nsIScriptContext>::get() line 631 + 3 bytes
nsCOMPtr<nsIScriptContext>::operator nsDerivedSafe<nsIScriptContext> *() line 644
GlobalWindowImpl::DropTimeout(nsTimeoutImpl * 0x05ae3c50, nsIScriptContext *
0x00000000) line 3568 + 11 bytes
nsGlobalWindow_RunTimeout(nsITimer * 0x05ae3b30, void * 0x05ae3c50) line 3717
nsTimer::Fire() line 194 + 17 bytes
nsTimerManager::FireNextReadyTimer(nsTimerManager * const 0x012cb430, unsigned
int 0) line 117
nsAppShell::Run(nsAppShell * const 0x00573c90) line 118
nsAppShellService::Run(nsAppShellService * const 0x0055da00) line 426
main1(int 1, char * * 0x00484260, nsISupports * 0x00000000) line 1234 + 32 bytes
main(int 1, char * * 0x00484260) line 1538 + 37 bytes
mainCRTStartup() line 338 + 17 bytes
KERNEL32! 77e97d08()

Comment 16

[Suresh]

I couldn't duplicate this bug using 7-20 and 7-18 BRANCH commercial bits in Win 2K.

Comment 17

[[e-mail] (gone - use [e-mail] instead)]

Hmm..that stack looks like http://bugzilla.mozilla.org/show_bug.cgi?id=91950 and
it's derivatives.

Comment 18

[sol]

Not sure if this bug is a dupe of bug 91950. I was not able to reproduce this
bug (91996) using the 2001-07-24 commercial branch on Win98. (I am also unable
to reproduce bug 91950 using 2001-07-24 commercial branch, but I did experience
the bug using 2001-07-23.)

Comment 19

[selmer (gone)]

Are we verified this is fixed on today's build?  Need to know ASAP!

Comment 20

[[e-mail] (gone - use [e-mail] instead)]

Whatever the fix/backout was, this seems to be working for me now with
2001-07-24-06-0.9.2 on Windows 2000.  Suresh, same for you?  If so, mark this
and then I'll verify it.

Comment 21

[Suresh]

This one seems to be fixed for me too using today's branch comm bits on win 2K.
Marking fixed. 

Comment 22

[[e-mail] (gone - use [e-mail] instead)]

verified.  I will re-file if I see this again (crossing my fingers).

Comment 23

[Johnny Stenback (:jst)]

I bet this was fixed by my fix for bug 91919.