540137 - Crash/hang at Japanese Flash using site

Status: RESOLVED INCOMPLETE

(Firefox :: General, defect)

Description

[Matthew Cline]

Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.3a1pre) Gecko/20100105 Minefield/3.7a1pre ID:20100105030950

Using Flash player beta 10.1 (full version 10.1.51.66)

At the given URL I've gotten three crash, which I've been unable to figure out how to reproduce.  The crash reports:

bp-b04b2733-1c56-48f3-ae14-e45672100115
bp-742dadc0-dde3-48c0-8a39-74ddf2100115
bp-cc37fc12-eb55-4bec-b139-1f0b72100115

I can reliably reproduce the browser hanging with 0% CPU usage: rapidly open the same URL in multiple new tabs by doing "Ctrl-middle-click" over and over, and Firefox will hang eventually.  Doing this without Flash installed results in no hang.  I killed the hung Firefox with "killall -SEGV firefox-bin" to get a crash report:

bp-ce95cc91-159b-4bfa-8368-647592100115

Comment 1

[timeless]

https://developer.mozilla.org/En/How_to_get_a_stacktrace_for_a_bug_report#Linux

Comment 2

[u279076]

Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US; rv:1.9.2.3pre) Gecko/20100405 Firefox/3.6.3plugin1

I'm finding this happen a lot when switching back to a tab with listen.grooveshark.com loaded after a while of listening to music on the site and browsing elsewhere.

http://crash-stats.mozilla.com/report/index/bp-06e5ea78-3585-4b48-81b4-832392100406
http://crash-stats.mozilla.com/report/index/bp-d49f896e-158b-40cd-b240-433042100406

I'll see if I can get a stacktrace next time.

Comment 3

[Karl Tomlinson (:karlt)]

Anthony, that might be related to bug 540197 (or more specifically https://bugzilla.gnome.org/show_bug.cgi?id=598476).  Which Ubuntu libgtk2.0-0 package do you have installed?

Comment 4

[u279076]

(In reply to comment #3)
> Anthony, that might be related to bug 540197 (or more specifically
> https://bugzilla.gnome.org/show_bug.cgi?id=598476).  Which Ubuntu libgtk2.0-0
> package do you have installed?

libgtk2.0.0 2.18.3-1ubuntu2.2

Comment 5

[Karl Tomlinson (:karlt)]

Thanks, Anthony.  I guess that's not bug 540197 then.
Your issue deserves a new bug report.  If you can reproduce this semi-reliably, can you try running firefox with --sync, please?
It might also be useful to know whether the crash happens with GDK_NATIVE_WINDOWS=1 in the environment.

The signature [@ linux-gate.so@0x430 ] pretty much just means that the process entered abort() or raise() on Ubuntu karmic, and so can be caused by many reasons.

[@ libxul.so@0x489c49 ] was a signature due to a problem with the crash report analysis that has since been resolved.

I'll remove both these signatures from the summary.

Matthew, if you see this your again, can you post some new crash report links, please?

Comment 6

[Wayne Mery (:wsmwk)]

anyone still see this problem?

Comment 7

[Phoenix]

Resolved per whiteboard