55809 - Browser crashes when saving file.

Status: VERIFIED DUPLICATE of bug 57326

(Core :: Networking, defect, P3)

Description

[kinmoz]

Using my 10/08/00 Win32 Netscape_20000922_BRANCH Mozilla debug build ...

If I load the following URL into my browser window:

    http://bugzilla.mozilla.org/showattachment.cgi?attach_id=16222

Then select 'File->Save As' from the menu, I get an assertion in 
AsyncReadStreamAdaptor::Cancel() because mEventQueueStreamListener is null.

If I continue past the assertion, I crash.

Here's the stack trace to the assertion I'm getting:


NTDLL! 77f762e8()
nsDebug::Assertion(const char * 0x01dd3454 
??_C@_0DJ@KMGL@You?5can?8t?5dereference?5a?5NULL?5nsC@, const char * 0x01dd3498 
??_C@_0N@NHHF@mRawPtr?5?$CB?$DN?50?$AA@, const char * 0x01dd39dc 
??_C@_0CB@MCGB@?4?4?2?4?4?2?4?4?2dist?2include?2nsCOMPtr?4h@, int 649) line 256 
+ 13 bytes
nsDebug::PreCondition(const char * 0x01dd3454 
??_C@_0DJ@KMGL@You?5can?8t?5dereference?5a?5NULL?5nsC@, const char * 0x01dd3498 
??_C@_0N@NHHF@mRawPtr?5?$CB?$DN?50?$AA@, const char * 0x01dd39dc 
??_C@_0CB@MCGB@?4?4?2?4?4?2?4?4?2dist?2include?2nsCOMPtr?4h@, int 649) line 396 
+ 21 bytes
nsCOMPtr<nsIStreamListener>::operator->() line 649 + 34 bytes
AsyncReadStreamAdaptor::Cancel(unsigned int 2153185313) line 78 + 36 bytes
nsMemCacheChannel::Cancel(nsMemCacheChannel * const 0x06568310, unsigned int 
2153185313) line 429
nsStreamListenerEvent::HandlePLEvent(PLEvent * 0x066df410) line 107 + 22 bytes
PL_HandleEvent(PLEvent * 0x066df410) line 580 + 10 bytes
PL_ProcessPendingEvents(PLEventQueue * 0x00abf2e0) line 513 + 9 bytes
_md_EventReceiverProc(HWND__ * 0x1a6c0274, unsigned int 49341, unsigned int 0, 
long 11268832) line 1049 + 9 bytes
USER32! 77e7124c()
00abf2e0()

Comment 1

[kinmoz]

Adding keywords: rtm, regression, dogfood, crash

Comment 2

[Gagan]

neeti: the stack suggests a dup of your cache crashers. I do not believe that 
its crashing for ALL download's and so am making this a dogfood- but we should 
consider its severity for rtm.

Comment 3

[neeti]

I am not able to reproduce this crash.

Comment 4

[neeti]

Kin,

I am not able to reproduce this crash. I notice that the stack trace attached 
has nsMemCacheChannel in it. If we load the browser, with the default settings 
for cache in the prefs panel, we should be using the disk cache, and not hitting 
the memory cache at all. What are your settings in the prefs panel for the 
cache, when you get this crash?

Thanks,
Neeti

Comment 5

[kinmoz]

I'm still seeing this crash with my Netscape_20000922_BRANCH debug build on 
Win32.

My prefs are set as follows:

    Memory Cache: 1024 KBytes
    Disk Cache:   7680 KBytes
    Compare the page in the cache to the page on the network:  "Once per 
session"

Comment 6

[kinmoz]

I doubt this matters, but I'm connected to Netscape via a 37,300 bps modem 
connection running SERA.

Comment 7

[neeti]

I am still not able to reproduce this with the latest branch build. Tever, can 
you help me in reproducing the crash?

Thanks,
Neeti

Comment 8

[neeti]

kin,

Do you crash, if you start with a clean cache directory?

Thanks,
Neeti

Comment 9

[kinmoz]

I'll give it a try ... the crash is happening on my home machine, and I'm not
seeing the crash here at Netscape.

Comment 10

[Gagan]

smaller visibility, hard to track => rtm-

Comment 11

[neeti]

I am marking this bug as duplicate of bug 57326

*** This bug has been marked as a duplicate of 57326 ***

Comment 12

[Tom Everingham]

verified dup