Closed Bug 29489 Opened 25 years ago Closed 25 years ago

Mozilla crashes on start up with Windows 2000 in NECKO.DLL

Categories

(Core :: Networking, defect, P3)

x86
Windows 2000
defect

Tracking

()

VERIFIED WORKSFORME

People

(Reporter: mozilla, Assigned: gagan)

Details

(Keywords: crash, relnote, Whiteboard: [PDT-])

This does not happen everytime, but it is frequent. I cannot come up with the exact necessary items to gaurantee a crash everytime, but here's how it happens when it does happen. 1) Open Mozilla and restore it from the taskbar (always open minimized on Windows 2000). 2) Click a couple times in the URL bar to enter a URL. 3) Start typing. 4) Crash. Crash is not captured by Dr. Watson, but Even Viewer has this to say: Event Type: Information Event Source: Application Popup Event Category: None Event ID: 26 Date: 02-28-00 Time: 01:57:07 User: N/A Computer: JERRY Description: Application popup: mozilla.org - Mozilla: mozilla.exe - Application Error : The instruction at "0x605691ec" referenced memory at "0x0187e000". The memory could not be "read". Click on OK to terminate the program Click on CANCEL to debug the program This is with build 2000022508
Sorry for the SPAM, setting CRASH keyword.
Keywords: crash
shrir, can you go into the lab and check this with Win2000? thanks!
QA Contact: paulmac → shrir
sure, I'll check with today's build and comment. Thnx !
It's probably not regular enough to see by playing for a few minutes. I can go for 20 starts or more and never see it. Sometimes I get it three times in a row. It will probably require someone that uses Mozilla on a daily basis (I do not) on Win2k to get better details.
So I'm no Shrir, but I gave this a few tries with the 2000022809 build on Win2k Professional Preview release 1. No luck. I didn't see the minimized on startup behavior either.
If it makes a difference, I am on the final, build 2195, of Advanced Server.
OK. It's not pressing keys that triggers the crash. It just appeared to be because I usually type in a URL at startup. It just crashes when started at random. It is always an access violation with necko.dll. Here: Event Type: Information Event Source: Application Popup Event Category: None Event ID: 26 Date: 02-28-00 Time: 11:19:34 User: N/A Computer: JERRY Description: Application popup: mozilla.org at a glance - Mozilla: mozilla.exe - Application Error : The instruction at "0x605691ec" referenced memory at "0x0193e000". The memory could not be "read". --------- 605691DD add ebx,2 605691E0 mov dword ptr [ebp-8],eax 605691E3 mov al,byte ptr [ebx-1] 605691E6 mov cl,byte ptr [ebx-2] 605691E9 mov byte ptr [ebp-2],al *605691EC mov al,byte ptr [ebx] 605691EE mov byte ptr [ebp-4],al 605691F1 inc ebx 605691F2 movsx eax,word ptr [ebp+0Ch] 605691F6 movzx edi,cl 605691F9 mov byte ptr [ebp+0Bh],cl 605691FC test dword ptr [edi*4+6056FC5Ch],eax ------------- I guess I need to learn how to build debug versions for this stuff.
Changing summary field
Summary: Typing in URL bar crashes on Windows 2000 → Mozilla crashes on start up with Windows 2000 in NECKO.DLL
Jerry - Please got get todays Win32 buld at: ftp://ftp.mozilla.org/pub/mozilla/nightly/latest/ Let us know if this happens with TODAY's build on the system you have. Delete all previous versions of mozilla and start with a fresh install. Thanks!
changing component to networking, reassigning.
Assignee: trudelle → gagan
Component: XP Toolkit/Widgets → Networking
QA Contact: shrir → tever
Still happens sometimes while loading a URL with build 2000022808. Here is the info I can give: Event Type: Information Event Source: Application Popup Event Category: None Event ID: 26 Date: 02-28-00 Time: 12:52:24 User: N/A Computer: JERRY Description: Application popup: mozilla.exe - Application Error : The instruction at "0x605791f4" referenced memory at "0x01d23000". The memory could not be "read". ---------- 605791D4 test eax,eax 605791D6 mov word ptr [ebp-2],cx 605791DA mov word ptr [ebp-4],cx 605791DE jle 6057929A 605791E4 push edi 605791E5 add ebx,2 605791E8 mov dword ptr [ebp-8],eax 605791EB mov al,byte ptr [ebx-1] 605791EE mov cl,byte ptr [ebx-2] 605791F1 mov byte ptr [ebp-2],al *605791F4 mov al,byte ptr [ebx] 605791F6 mov byte ptr [ebp-4],al 605791F9 inc ebx 605791FA movsx eax,word ptr [ebp+0Ch] 605791FE movzx edi,cl 60579201 mov byte ptr [ebp+0Bh],cl 60579204 test dword ptr [edi*4+6057FC4Ch],eax 6057920B jne 60579270 6057920D cmp cl,25h 60579210 jne 60579248 60579212 lea eax,[ebp-2] ----------
Forgive me if this is not what we need, but I built a debug version and got it to crash. Here's what I got out of the call stack: NTDLL! 77f9f9df() nsDebug::Assertion(const char * 0x021dcee0, const char * 0x021dceb8, const char * 0x021dce7c, int 580) line 189 + 13 bytes nsDebug::WarnIfFalse(const char * 0x021dcee0, const char * 0x021dceb8, const char * 0x021dce7c, int 580) line 247 + 21 bytes nsURILoader::Stop(nsURILoader * const 0x02426070, nsISupports * 0x0356e2e0) line 580 + 58 bytes nsDocShell::StopLoad(nsDocShell * const 0x0356e920) line 191 nsWebShell::StopLoad(nsWebShell * const 0x0356e920) line 3781 nsWebShell::StopBeforeRequestingURL(nsWebShell * const 0x0356e9e0) line 2203 nsWebShell::DoLoadURL(nsIURI * 0x03580160, const char * 0x0038215c, nsIInputStream * 0x00000000, unsigned int 10, const unsigned int 0, const unsigned short * 0x0012d818, const char * 0x00000000, int 1) line 1560 nsWebShell::LoadURI(nsWebShell * const 0x0356e9e0, nsIURI * 0x03580160, const char * 0x0038215c, nsIInputStream * 0x00000000, int 0, unsigned int 10, const unsigned int 0, nsISupports * 0x035a4960, const unsigned short * 0x0012d818, const char * 0x00000000) line 1959 + 44 bytes nsWebShell::LoadURL(nsWebShell * const 0x0356e9e0, const unsigned short * 0x035803e0, const char * 0x0038215c, nsIInputStream * 0x00000000, int 0, unsigned int 10, const unsigned int 0, nsISupports * 0x035a4960, const unsigned short * 0x0012d818, const char * 0x00000000) line 2183 + 53 bytes nsWebShell::LoadURL(nsWebShell * const 0x0356e9e0, const unsigned short * 0x035803e0, nsIInputStream * 0x00000000, int 0, unsigned int 10, const unsigned int 0, nsISupports * 0x035a4960, const unsigned short * 0x0012d818) line 1402 nsHistoryEntry::Load(nsIWebShell * 0x0356e9e0, int 0) line 580 nsSessionHistory::Goto(nsSessionHistory * const 0x03607290, int 1, nsIWebShell * 0x0356e9e0, int 0) line 1129 + 16 bytes nsSessionHistory::GoForward(nsSessionHistory * const 0x03607290, nsIWebShell * 0x0356e9e0) line 1171 nsBrowserInstance::GoForward(nsBrowserInstance * const 0x036073bc, nsIWebShell * 0x0356e9e0) line 1816 + 22 bytes nsBrowserInstance::Forward(nsBrowserInstance * const 0x036073b0) line 315 XPTC_InvokeByIndex(nsISupports * 0x036073b0, unsigned int 4, unsigned int 0, nsXPTCVariant * 0x0012dbf0) line 139 nsXPCWrappedNativeClass::CallWrappedMethod(JSContext * 0x031d2cf0, nsXPCWrappedNative * 0x035a4180, const XPCNativeMemberDescriptor * 0x035a5948, nsXPCWrappedNativeClass::CallMode CALL_METHOD, unsigned int 0, long * 0x02f458d8, long * 0x0012ddb0) line 898 + 43 bytes WrappedNative_CallMethod(JSContext * 0x031d2cf0, JSObject * 0x00e18490, unsigned int 0, long * 0x02f458d8, long * 0x0012ddb0) line 200 + 34 bytes js_Invoke(JSContext * 0x031d2cf0, unsigned int 0, unsigned int 0) line 665 + 26 bytes js_Interpret(JSContext * 0x031d2cf0, long * 0x0012e69c) line 2292 + 15 bytes js_Invoke(JSContext * 0x031d2cf0, unsigned int 0, unsigned int 0) line 681 + 13 bytes js_Interpret(JSContext * 0x031d2cf0, long * 0x0012ef44) line 2292 + 15 bytes js_Invoke(JSContext * 0x031d2cf0, unsigned int 1, unsigned int 2) line 681 + 13 bytes js_InternalInvoke(JSContext * 0x031d2cf0, JSObject * 0x00e17940, long 14776656, unsigned int 0, unsigned int 1, long * 0x0012f0d0, long * 0x0012f07c) line 754 + 19 bytes JS_CallFunctionValue(JSContext * 0x031d2cf0, JSObject * 0x00e17940, long 14776656, unsigned int 1, long * 0x0012f0d0, long * 0x0012f07c) line 2790 + 31 bytes nsJSContext::CallEventHandler(nsJSContext * const 0x031d6d60, void * 0x00e17940, void * 0x00e17950, unsigned int 1, void * 0x0012f0d0, int * 0x0012f0cc) line 562 + 33 bytes nsJSEventListener::HandleEvent(nsIDOMEvent * 0x03583f44) line 128 + 57 bytes nsEventListenerManager::HandleEventSubType(nsListenerStruct * 0x034f4de0, nsIDOMEvent * 0x03583f44, unsigned int 8, unsigned int 7) line 697 + 19 bytes nsEventListenerManager::HandleEvent(nsIPresContext * 0x031f1030, nsEvent * 0x0012f5f8, nsIDOMEvent * * 0x0012f5cc, unsigned int 7, nsEventStatus * 0x0012f638) line 1456 + 35 bytes nsXULElement::HandleDOMEvent(nsXULElement * const 0x0336ea00, nsIPresContext * 0x031f1030, nsEvent * 0x0012f5f8, nsIDOMEvent * * 0x0012f5cc, unsigned int 1, nsEventStatus * 0x0012f638) line 3079 nsTitledButtonFrame::MouseClicked(nsIPresContext * 0x031f1030) line 1327 nsTitledButtonFrame::HandleEvent(nsTitledButtonFrame * const 0x00e10d3c, nsIPresContext * 0x031f1030, nsGUIEvent * 0x0012f684, nsEventStatus * 0x0012f948) line 1232 nsEventStateManager::CheckForAndDispatchClick(nsEventStateManager * const 0x03510260, nsIPresContext * 0x031f1030, nsMouseEvent * 0x0012fa3c, nsEventStatus * 0x0012f948) line 1711 + 30 bytes nsEventStateManager::PostHandleEvent(nsEventStateManager * const 0x03510260, nsIPresContext * 0x031f1030, nsGUIEvent * 0x0012fa3c, nsIFrame * 0x00e10d3c, nsEventStatus * 0x0012f948, nsIView * 0x031f2450) line 889 + 24 bytes PresShell::HandleEvent(PresShell * const 0x031f2e64, nsIView * 0x031f2450, nsGUIEvent * 0x0012fa3c, nsEventStatus * 0x0012f948) line 2959 + 43 bytes nsView::HandleEvent(nsView * const 0x031f2450, nsGUIEvent * 0x0012fa3c, unsigned int 28, nsEventStatus * 0x0012f948, int & 0) line 799 nsViewManager2::DispatchEvent(nsViewManager2 * const 0x031f37c0, nsGUIEvent * 0x0012fa3c, nsEventStatus * 0x0012f948) line 1216 HandleEvent(nsGUIEvent * 0x0012fa3c) line 69 nsWindow::DispatchEvent(nsWindow * const 0x031f2334, nsGUIEvent * 0x0012fa3c, nsEventStatus & nsEventStatus_eIgnore) line 493 + 10 bytes nsWindow::DispatchWindowEvent(nsGUIEvent * 0x0012fa3c) line 514 nsWindow::DispatchMouseEvent(unsigned int 301, nsPoint * 0x00000000) line 2957 + 21 bytes ChildWindow::DispatchMouseEvent(unsigned int 301, nsPoint * 0x00000000) line 3175 nsWindow::ProcessMessage(unsigned int 514, unsigned int 0, long 3932241, long * 0x0012fcd8) line 2243 + 24 bytes nsWindow::WindowProc(HWND__ * 0x000606e0, unsigned int 514, unsigned int 0, long 3932241) line 671 + 27 bytes USER32! 77e13eb0() USER32! 77e1401a() USER32! 77e192da() nsAppShellService::Run(nsAppShellService * const 0x023f2eb0) line 400 main1(int 1, char * * 0x00b34bc0, nsISplashScreen * 0x00000000) line 763 + 32 bytes main(int 1, char * * 0x00b34bc0) line 883 + 17 bytes mainCRTStartup() line 338 + 17 bytes KERNEL32! 77e87903()
I'm seeing this on Win2k as well. Once it crashed while typing in the proxy server in the HTTP field. Other times is just crashing all over the place. Only in the last few daily builds.
Seems like a beta candidate. Do we care about windows 2000? Rel-note?
Keywords: beta1
Target Milestone: M14
I can tell you that I am unable to reproduce this crash in debug builds. In the optimized builds it is pretty frequent and occurs immediately at launch.
Putting on PDT+ for beta1. But will move to PDT- if not fixed by 03/03 or if this regresses. cing gbush, I will get you Win200 final CD.
Whiteboard: [PDT+] w/b minus on 03/03
QA Contact: tever → gbush
So far we haven't been able replicate this (thanks cathleen). Could other people who have W2K try this and confirm? cc'ing troy. thanks
unable to reproduce on sean's m/c as well...(thanks cathleen) marking WORKSFORME
Status: NEW → RESOLVED
Closed: 25 years ago
Resolution: --- → WORKSFORME
I want to make sure that you guys are trying to verify this with an optimized build, and on a Windows 2000 final machine. That is the case right?
yes we've tried with the 2195 build of Win2k(with 2000030209 seamonkey). I think it might have something to do with different install options. I note where you say you're using Win2k(Advanced Server). We still have to look into that. phils@vic.mondial.net.au, do you still see this crash? at startup? what version of Win2k are you using?
I have done a fair bit of viewing with build 2000030116 and haven't had a crash yet. So looks like it's fixed. This is with win2k professional (normal workstation version).
jerry - please try the latest build at: for M14 final - http://www.mozilla.org/projects/seamonkey/release-notes/ latest nightly daily - http://www.mozilla.org/binaries.html Marking Verified/WorksForMe
Status: RESOLVED → VERIFIED
I don't know the differences between the two, but I note that everyone(including myself) who has claimed to give this a go and not seen a crash reported using Win2k Pro - whereas jerry reports his crash using Win2k Server. Maybe that's the crucial difference?
Status: VERIFIED → REOPENED
Resolution: WORKSFORME → ---
I have not seen the crash in M14 final or in any of the builds I have pulled and made from CVS today. I will wait another day and then mark WORKSFORME if it's still OK.
I've extended the date of transition to PDT- to allow this bug to go away as resolved (works4me).
Whiteboard: [PDT+] w/b minus on 03/03 → [PDT+] w/b minus on 3/5
Marking worksforme, per comment of several days ago.
Status: REOPENED → RESOLVED
Closed: 25 years ago25 years ago
Resolution: --- → WORKSFORME
I thought it had already been resolved for that by Gagan or I would've came back and marked it myself. Sorry.
Marking Verified.
Status: RESOLVED → VERIFIED
It's back. Using 2000030808. The call stack with no symbols (doesn't happen in debug) is: NECKO! 6057d115() NECKO! 6057cc52() e8505660()
Status: VERIFIED → REOPENED
Resolution: WORKSFORME → ---
This is a bothersome regression... but I'm clearing the PDT field since we're well past the due date. This will allow the PDT to decide if we need to hold beta1 for this, or to release note the problem. Can folks comment on whether this is an intermittent bug, or a consistent bug? Does it happen on "all" Win 2000 boxes? Thanks, Jim
Whiteboard: [PDT+] w/b minus on 3/5
Severity: normal → critical
Putting on PDT- radar for beta1. Will release note if problems with Win2000. gbush, do you see this problem with latest builds?
Keywords: relnote
Whiteboard: [PDT-]
beta branch build 2000031506 on win2000 lab machine installed and launched and ran successfully- not able to force a crash using suggestions here
Moving to M15
Target Milestone: M14 → M15
jerrybaker@weirdness.com: is this still happening?
Target Milestone: M15 → M16
To be honest, I haven't downloaded or built Mozilla since a couple weeks ago. I will give it a go in the next few days. If you don't hear from me, then it would be safe to say that it no longer happens.
Moving to M17 which is now considered part of beta2.
Target Milestone: M16 → M17
Haven't heard from the reporter. Marking as WORKSFORME again...
Status: REOPENED → RESOLVED
Closed: 25 years ago25 years ago
Resolution: --- → WORKSFORME
ok
Status: RESOLVED → VERIFIED
Mass removing self from CC list.
Now I feel sumb because I have to add back. Sorry for the spam.
You need to log in before you can comment on or make changes to this bug.