Closed Bug 78317 Opened 24 years ago Closed 24 years ago

Crash at startup

Categories

(SeaMonkey :: Installer, defect)

x86
Windows NT
defect
Not set
blocker

Tracking

(Not tracked)

VERIFIED FIXED

People

(Reporter: a.schild, Assigned: leaf)

References

Details

(Keywords: crash, smoketest, topcrash)

Attachments

(1 file)

From Bugzilla Helper: User-Agent: Mozilla/4.75 [de] (Windows NT 5.0; U) BuildID: 200104301500 On W2K Mozilla crashes on startup with the build from yesterday 20010430 builds from 15:00 and 23:00. Reproducible: Always Steps to Reproduce: Install Mozilla and start it with either a Mozilla Modern theme profile or migrate a existing 4.x profile I tried a completely clean installation and Mozilla only starts when I say to create a new profile. When either migrating a 4.x profile, or using my (Modern theme) Mozilla Profile (I used for now ~4 months) then Mozilla just crashes on startup
TB are TB29846851K TB29846686K and TB29846680W
Same problem on Win98SE. I had changed my chrome to Modern. Downloaded installer from 30-Apr-2001 23:34. (Couldn't get far enough in to get build #). When I then installed 2001042908 back over it, it requested that I choose a profile and changed my chrome back to Classic (in case that tells you anything about whether it's a profile problem). Here is the error message: MOZILLA caused an invalid page fault in module GKLAYOUT.DLL at 0167:6034c2e1. Registers: EAX=00000000 CS=0167 EIP=6034c2e1 EFLGS=00010246 EBX=01399090 SS=016f ESP=0068f334 EBP=0068f408 ECX=01361c54 DS=016f ESI=00000001 FS=445f EDX=0068f3e0 ES=016f EDI=0138a560 GS=0000 Bytes at CS:EIP: 8b 08 50 ff 51 28 8b 45 f0 57 2b 45 e4 8b cb 89 Stack dump: 0068f3e0 00c7fe58 0068f450 00c7fe58 00000000 53af834c 00000000 00000246 000083c6 0068f3aa bff7168f 01572737 53af0000 01572737 00000000 00000000
also seen on commercial build 2001-05-01-06-trunk adding smoketest and crash keywords
Status: UNCONFIRMED → NEW
Ever confirmed: true
Keywords: crash, smoketest
I installed the new build 2001.05.01 time 7:32 and it still crashed. Then I did step back to 2001043004 and now I'm able to use my profile with the new modern theme...... André.
cc'ing smoketester (myself)
does using the classic theme crash in the same way? could this be related to the modern landing? cc'ing hangas
This crashes for me with a new Profile (using Classic theme) as well (on Win2k).
Moving to layout for a guess.
Assignee: asa → karnaze
Component: Browser-General → Layout
QA Contact: doronr → petersen
this is also happening on mfcEmbed...although it manages to get the browser up before crashing
*** Bug 78370 has been marked as a duplicate of this bug. ***
I experienced this crash with a clean fresh profile-less 23:00 win32-installer build. Verified that the 16:00 installer worked with a clean and fresh install, then went back to the 2300 win32-talkback.zip build without a profile which worked as well, atleast for me.
I'm seeing this crash on Win98 before it even gets to the profile manager.
Here is my call stack nsGfxScrollFrameInner::Layout [d:\builds\seamonkey\mozilla\layout\html\base\src\nsGfxScrollFrame.cpp, line 1258] nsGfxScrollFrame::DoLayout [d:\builds\seamonkey\mozilla\layout\html\base\src\nsGfxScrollFrame.cpp, line 1047] nsBox::Layout [d:\builds\seamonkey\mozilla\layout\xul\base\src\nsBox.cpp, line 985] nsBoxFrame::Reflow [d:\builds\seamonkey\mozilla\layout\xul\base\src\nsBoxFrame.cpp, line 781] nsGfxScrollFrame::Reflow [d:\builds\seamonkey\mozilla\layout\html\base\src\nsGfxScrollFrame.cpp, line 738] nsContainerFrame::ReflowChild [d:\builds\seamonkey\mozilla\layout\html\base\src\nsContainerFrame.cpp, line 745] ViewportFrame::Reflow [d:\builds\seamonkey\mozilla\layout\html\base\src\nsViewportFrame.cpp, line 538] PresShell::InitialReflow [d:\builds\seamonkey\mozilla\layout\html\base\src\nsPresShell.cpp, line 2577] HTMLContentSink::StartLayout [d:\builds\seamonkey\mozilla\content\html\document\src\nsHTMLContentSink.cpp, line 3577] HTMLContentSink::OpenBody [d:\builds\seamonkey\mozilla\content\html\document\src\nsHTMLContentSink.cpp, line 2946] CNavDTD::OpenBody [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 3123] CNavDTD::OpenContainer [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 3374] CNavDTD::HandleDefaultStartToken [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 1328] CNavDTD::HandleStartToken [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 1733] CNavDTD::HandleToken [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 913] CNavDTD::BuildModel [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 553] nsParser::BuildModel [d:\builds\seamonkey\mozilla\htmlparser\src\nsParser.cpp, line 1985] nsParser::ResumeParse [d:\builds\seamonkey\mozilla\htmlparser\src\nsParser.cpp, line 1864] nsParser::OnStopRequest [d:\builds\seamonkey\mozilla\htmlparser\src\nsParser.cpp, line 2344] nsDocumentOpenInfo::OnStopRequest [d:\builds\seamonkey\mozilla\uriloader\base\nsURILoader.cpp, line 277] nsStreamIOChannel::OnStopRequest [d:\builds\seamonkey\mozilla\netwerk\base\src\nsInputStreamChannel.cpp, line 461] nsOnStopRequestEvent::HandleEvent [d:\builds\seamonkey\mozilla\netwerk\base\src\nsRequestObserverProxy.cpp, line 159] PL_HandleEvent [d:\builds\seamonkey\mozilla\xpcom\threads\plevent.c, line 589] PL_ProcessPendingEvents [d:\builds\seamonkey\mozilla\xpcom\threads\plevent.c, line 522] _md_EventReceiverProc [d:\builds\seamonkey\mozilla\xpcom\threads\plevent.c, line 1070] nsAppShellService::Run [d:\builds\seamonkey\mozilla\xpfe\appshell\src\nsAppShellService.cpp, line 408] netscp6.exe + 0x16eb (0x004016eb) netscp6.exe + 0x11b8 (0x004011b8) netscp6.exe + 0x2b92 (0x00402b92) KERNEL32.DLL + 0x192a6 (0x77e992a6)
Is this windows-only? I'm not seeing anything like this on my Linux build from last night.
Brian, yes, this is Windows only...linux and mac builds this morning were fine
http://bonsai.mozilla.org/cvsquery.cgi?module=MozillaTinderboxAll&branch=HEAD&cv sroot=/cvsroot&date=explicit&mindate=988684291&maxdate=988685820&who=rbs%25maths .uq.edu.au might be the culprit... rbs, can you take this bug?
Pulling, then building, then investigating... I didn't have the new modern theme at the time of my tests. Will report any findings as they become available.
i built win98SE build locally after the close - works fine , no crash in either modern or classic
Crash seen on Win NT 2001-05-01-06-trunk (commercial). Even after installation and trying to restart, it crashes (access violation). No profile gets created. Cannot run Profile Manager either.
with a pre-existing profile? the release builds are what is showing this behaviour, and they were built post-closure.
My build isn't finished yet. But from my pre-checkin tests, the most plausible cause for the crash could just be a packaging issue. I added a critical file 'fontEncoding.properties' and I know for sure that a crash will happen if that file is missing. (In a debug build, an assertion is first fired, but in an optimized, it goes straight to the crash.) Is that possible that this file was left out from the packaging process?
d'oh. where does the file go in dist? if it goes someplace that isn't recursively covered in http://lxr.mozilla.org/seamonkey/source/xpinstall/packager/packages-win it's getting left out. That's why, at the top of tinderbox, it describes the need to update that file if you add critical new files. <homer>d'oh!</homer> I'll fix and repackage.
Here's my stab at a fix... i'm assuming that other font stuff is going to have to be packaged, right? Index: packages-win =================================================================== RCS file: /cvsroot/mozilla/xpinstall/packager/packages-win,v retrieving revision 1.147 diff -u -r1.147 packages-win --- packages-win 2001/04/17 14:33:15 1.147 +++ packages-win 2001/05/01 20:44:41 @@ -252,6 +252,7 @@ bin\res\viewsource.css bin\res\arrow.gif bin\res\html\* +bin\res\fonts\* bin\defaults\wallet\VcardSchema.tbl bin\defaults\wallet\FieldSchema.tbl bin\defaults\wallet\SchemaConcat.tbl
if that's all this takes, then sr=alecf
Sorry about that.
Yes, all the 'fonts\*' are needed. Otherwise rendering is problematic... r=rbs
fix committed, reassigning to me.
Assignee: karnaze → leaf
Component: Layout → Installer: XPI Packages
Still seeing this in Current 1 May Windows installer build (in case you've rebuilt it already) at 5:15 pm EDST. MOZILLA caused an invalid page fault in module GKLAYOUT.DLL at 0167:6034c301. etc.
My build has completed and I am not seeing the crash.
fix committed, builds respun.
Status: NEW → RESOLVED
Closed: 24 years ago
Resolution: --- → FIXED
Fixed in Windows installer build 2001050204. Thanks, folks.
verified on windows commercial build 2001-05-02-12-trunk
Status: RESOLVED → VERIFIED
This bug is a topcrash for M09. Added topcrash keyword. Here are some comments from recent crashes: (29873420) Comments: Installation of today's build from sweetlou (05/01/2001) was just finishing. (29870028) Comments: startup crash with 2001050106 build (29862673) Comments: bug already filed for this issue...bugzilla ID 78355 (29859092) Comments: launching today's build (29855967) Comments: startup (29855662) Comments: starting up (29848972) Comments: 1st startup after install (29847201) Comments: crash on startup after running installer (??) (29847199) Comments: crash on startup (after running installer). And here is a recent stack trace: Incident ID 29882861 nsGfxScrollFrameInner::Layout [d:\builds\seamonkey\mozilla\layout\html\base\src\nsGfxScrollFrame.cpp, line 1258] nsGfxScrollFrame::DoLayout [d:\builds\seamonkey\mozilla\layout\html\base\src\nsGfxScrollFrame.cpp, line 1047] nsBox::Layout [d:\builds\seamonkey\mozilla\layout\xul\base\src\nsBox.cpp, line 985] nsBoxFrame::Reflow [d:\builds\seamonkey\mozilla\layout\xul\base\src\nsBoxFrame.cpp, line 781] nsGfxScrollFrame::Reflow [d:\builds\seamonkey\mozilla\layout\html\base\src\nsGfxScrollFrame.cpp, line 738] nsContainerFrame::ReflowChild [d:\builds\seamonkey\mozilla\layout\html\base\src\nsContainerFrame.cpp, line 745] ViewportFrame::Reflow [d:\builds\seamonkey\mozilla\layout\html\base\src\nsViewportFrame.cpp, line 538] PresShell::InitialReflow [d:\builds\seamonkey\mozilla\layout\html\base\src\nsPresShell.cpp, line 2577] HTMLContentSink::StartLayout [d:\builds\seamonkey\mozilla\content\html\document\src\nsHTMLContentSink.cpp, line 3577] HTMLContentSink::OpenBody [d:\builds\seamonkey\mozilla\content\html\document\src\nsHTMLContentSink.cpp, line 2946] CNavDTD::OpenBody [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 3123] CNavDTD::OpenContainer [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 3374] CNavDTD::HandleDefaultStartToken [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 1328] CNavDTD::HandleStartToken [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 1733] CNavDTD::HandleToken [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 913] CNavDTD::BuildModel [d:\builds\seamonkey\mozilla\htmlparser\src\CNavDTD.cpp, line 553] nsParser::BuildModel [d:\builds\seamonkey\mozilla\htmlparser\src\nsParser.cpp, line 1985] nsParser::ResumeParse [d:\builds\seamonkey\mozilla\htmlparser\src\nsParser.cpp, line 1864] nsParser::OnStopRequest [d:\builds\seamonkey\mozilla\htmlparser\src\nsParser.cpp, line 2344] nsDocumentOpenInfo::OnStopRequest [d:\builds\seamonkey\mozilla\uriloader\base\nsURILoader.cpp, line 277] nsStreamIOChannel::OnStopRequest [d:\builds\seamonkey\mozilla\netwerk\base\src\nsInputStreamChannel.cpp, line 461] nsOnStopRequestEvent::HandleEvent [d:\builds\seamonkey\mozilla\netwerk\base\src\nsRequestObserverProxy.cpp, line 159] PL_HandleEvent [d:\builds\seamonkey\mozilla\xpcom\threads\plevent.c, line 589] PL_ProcessPendingEvents [d:\builds\seamonkey\mozilla\xpcom\threads\plevent.c, line 522] _md_EventReceiverProc [d:\builds\seamonkey\mozilla\xpcom\threads\plevent.c, line 1070] nsAppShellService::Run [d:\builds\seamonkey\mozilla\xpfe\appshell\src\nsAppShellService.cpp, line 408] Netscp6.exe + 0x16eb (0x004016eb) Netscp6.exe + 0x11b8 (0x004011b8) Netscp6.exe + 0x2b92 (0x00402b92) KERNEL32.DLL + 0x192a6 (0x77e992a6)
Status: VERIFIED → REOPENED
Keywords: topcrash
Resolution: FIXED → ---
> This bug is a topcrash for M09. How come? As it can be seen from the link in the "Additional Comments From Daniel (Leaf) Nunes 2001-05-01 12:17", the code under consideration here wasn't landed on the M0.9 branch.
If that's the case, then please note such things explicitly in the bug. (It will make life easier for everyone.) Thanks. janc
Not sure if this one is still on the radar? I pulled from the M0.9 branch and could compile and run without problems (on Win2K).
Well, trunk build 2001050320 runs just fine for me on Win2k and Win98FE. I can do some tests on NT4, Win95, and WinME later, if needed.
Thanks for the offer. But janc is referring to the branch. The trunk is okay.
OK, just tried 0.9 branch 2001050322. Crashes on both Win98FE and Win2000. I sent a talkback report, but I don't know how to get the report ID.
I was building with MathML enabled. I tried another build with MathML disabled and I still don't crash. It is puzzling. I am trying another colobber/re-build of an optimized build to see if there is any difference. On a lighter note, it is rather ironic that I am wishing to crash :-) How could one wants so dearly to crash!?!
My own optimized build completed and it worked fine too. The official .zip and installer.exe work too. My debug build work too. So I am at a lost as to what else to try. As noted before the startup crash at "nsGfxScrollFrameInner::Layout [d:\builds\seamonkey\mozilla\layout\html\base\src\nsGfxScrollFrame.cpp, line 1258]", at "fm->GetHeight(fontHeight)" where the crash is because "fm" is NULL, is likely due to the missing "res\fonts\*" when packaging but that should only happen on the m0.9+ trunk and the fix for that was successfully landed there sometime ago. Another bug should perhaps be open because the crash could be caused by a reason other than the earlier reason narrowed in this bug. This bug may be harmful now to m0.9 because it is having the side-effects of preventing a wider investigation suitable to m0.9.
re-resolving. If there is a problem on the branch, you need to file a new bug, but you probably want to check with drivers since I doubt any more fixes are going into 0.9 at this point.
Status: REOPENED → RESOLVED
Closed: 24 years ago24 years ago
Resolution: --- → FIXED
Marking works for me with Windows Me May 21 build.
Status: RESOLVED → VERIFIED
Product: Browser → Seamonkey
Component: Installer: XPI Packages → Installer
QA Contact: chrispetersen → general
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: