When visiting http://www.gamestar.de/aktuell/news/news6397.html with JavaScript enabled Mozilla just crashes. I tested this with current milestone 0.9.8 running on Linux, the old 0.9.7 made no problems.

When visiting http://www.gamestar.de/aktuell/news/news6397.html with JavaScript enabled Mozilla just crashes. I tested this with current milestone 0.9.8 running on Linux, the old 0.9.7 made no problems.

This bug is not reproduceable under Windows ME. -- Sorry for the double postings up there.

it hangs moz on 2002020412/linux. doesn't eat a lot of cpu - just hangs. Is this some gtk timing problem, since it works on win and stuff?

Sev: nor -> cri Key: +crash (WFM 2002020208/WinNT4)

WFM WinXP/2002020409

reporter, please try using a talkback release.

Back to Browser-General until we can get a stack trace; there is no reason to believe this is a crash in JS Engine. Mozilla releases nightly and milestone builds with Full Circle's Talkback. Talkback should catch most crashes and offer to send in a crash report. We can retrieve that crash report and attach it to your bug report if you provide either the "Incident ID" (you can get it by running the Talkback program from /components/talkback/) or you can let us know the email address you used to submit the report and the time of sending. Once we have a stack trace, we'll know which component to send this to. I don't crash on this with my trunk Linux build from 20020204xx. Thanks; cc'ing self -

Today Mozilla just froze when visiting the link, I had to kill it. Talkback send a report, ID TB2577400G I visited the site again, this time with JavaScript disabled. No crash. I turned JavaScript on and clicked on reload. No crash. I forced a complete reload with SHIFT + reload: The browser just crashed like I reported yesterday. Talkback ID TB2577522H For your information: My Linux system is newly installed Mandrake 8.1 without any changes.

Same behaviour with Milestone 0.9.7, sometimes a complete crash (the program ends suddenly), sometimes it only freezes and doesn't respond.

Philipp, thanks! Incident ID 2577400 Stack Signature libc.so.6 + 0x8118e (0x4056c18e) 19b3544a Trigger Time 2002-02-06 06:19:42 URL visited User Comments Build ID 2002020415 Product ID MozillaBranch Platform Operating System LinuxIntel Module Trigger Reason SIGSEGV: Segmentation Fault: (signal 11) Stack Trace libc.so.6 + 0x8118e (0x4056c18e) libc.so.6 + 0x80f44 (0x4056bf44) libX11.so.6 + 0x2ff0f (0x4040df0f) libgdk-1.2.so.0 + 0x2922c (0x4038d22c) nsRegionGTK::SetTo() nsWindow::Update() nsWindow::UpdateIdle() ----------------------------------------------------------------- Incident ID 2577522 Stack Signature nsSupportsArray::ElementAt() b37e89cc Trigger Time 2002-02-06 06:22:54 URL visited User Comments Build ID 2002020415 Product ID MozillaBranch Platform Operating System LinuxIntel Module Trigger Reason SIGSEGV: Segmentation Fault: (signal 11) Stack Trace nsSupportsArray::ElementAt() imgContainer::Notify() nsTimerImpl::Process() handleMyEvent() PL_HandleEvent() PL_ProcessPendingEvents() nsEventQueueImpl::ProcessPendingEvents() event_processor_callback() our_gdk_io_invoke()

In LXR, imgContainer::Notify is at: http://lxr.mozilla.org/seamonkey/source/modules/libpr0n/src/imgContainer.cpp#372 AND nsRegionGTK::SetTo is at: http://lxr.mozilla.org/seamonkey/source/gfx/src/gtk/nsRegionGTK.cpp#109 http://lxr.mozilla.org/seamonkey/source/gfx/src/gtk/nsRegionGTK.cpp#118 http://lxr.mozilla.org/seamonkey/source/gfx/src/gtk/nsRegionGTK2.cpp#67 http://lxr.mozilla.org/seamonkey/source/gfx/src/gtk/nsRegionGTK2.cpp#76 I'm not sure of the component for this, but it doesn't look like JS Engine. Myself, I do not hang or crash with my Linux builds. I don't know if this is relevant or not, but my debug build shows lots of these asserts: ###!!! Break: at file nsImageGTK.cpp, line 624 ###!!! ASSERTION: You can't draw an image with a 0 width or height!: 'aSWidth > 0 && aDWidth > 0 && aSHeight > 0 && aDHeight > 0', file nsImageGTK.cpp, line 624 Note the question in Comment #3 above: > It hangs moz on 2002020412/linux. doesn't eat a lot of cpu - just hangs. > Is this some gtk timing problem, since it works on win and stuff? Reassigning to ImageLib for further triage -

I made some tests: I took the source of the page and reduced it step by step until I had the following code which made mozilla crash: <html> <head> <title>http://www.gamestar.de/aktuell/news/news6397.html</title> </head> <body> <img src="http://ad.de.doubleclick.net/ad/idg.gamestar.de/aktuell;sz=120x60;ord=8208187906737174 "> </body> </html> This image isn't loaded in the original page with JavaScript enabled. In fact the image itself is enough to crash Mozilla. I've uploaded the image under http://www.newswriter.org/mozkiller.gif. Try loading this image with Mozilla under Linux. If everything seems alright press SHIFT + reload several times. On my machine Mozilla always crashes/freezes.

Philipp, thank you for reducing this! I will put it below as an attachment, just adding height="100", width="100" to the <IMG>. That will enable us to detect it more easily -

In both Mozilla and NN4.7, if I load the testcase and Shift + Reload it, I see the container for the image appear, remain on screen with the image still unresolved, and then the entire container disappears, leaving a blank page. I noticed that if you load the SRC for the image directly in the URL bar in any browser, it redirects: http://ad.de.doubleclick.net/ad/idg.gamestar.de/aktuell;sz=120x60;ord=8208187906 737174 redirects to http://ad.de.doubleclick.net/viewad/817-grey.gif I wonder if that has something to do with the problem? At any rate, I'm not hanging or crashing with my builds (WinNT 20020213xx, Linux 20020207xx).

Is this bug still valid? Not able to reproduce with Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2a) Gecko/20020910, pulled from cvs

Not crashing for me either on win2000/Linux. I tried with numerous versions moz1.0 -> 1.3 maybe the sight changed. will probably close soon as Works for Me unless someone says they are still seeing this.

I originaly reported this bug. The site itself changed, but in the disussion above we were able to track down the problem to one single image, which I accidently deleted some time ago, sorry. But I'm quite sure when I last tried to view this image with mozilla it worked alright, don't believe it still is a bug.

THANKS. If you happen to see it again let us know. Sorry that this took so long to get fixed...