Closed Bug 544153 Opened 15 years ago Closed 15 years ago

[OOPP] Crash (no Sig) when enabling flashblock and visiting a flash site with OOPP enabled

Categories

(Core Graveyard :: Plug-ins, defect)

x86
Linux
defect
Not set
critical

Tracking

(Not tracked)

VERIFIED DUPLICATE of bug 540114

People

(Reporter: tchung, Assigned: karlt)

References

Details

(Whiteboard: [OOPPTestday])

Attachments

(2 files, 1 obsolete file)

Attached image Flashblock crash screenshot (obsolete) (deleted) —
I am getting a crash (with a useless stacktrace), each time i enable flashblock, and visit a flash content site. The only message shown is a warning: "WARNING: Json file missing PluginName; iteration over non-sequence" Unhelpful Stack traces: http://crash-stats.mozilla.com/report/index/bp-293ecf31-5d34-42d8-a634-ff4c92100203 http://crash-stats.mozilla.com/report/index/bp-e58f2047-69b2-4386-a452-183262100203 Notes: When flashblock 1.5.11.2 is enabled and crashes the flash site, note that mozilla-runtime is never running. Also, if you test with dom.ipc.plugins.enabled == False and run the steps, you will NOT see the crash. Repro: 1) download nightly 2/03 build: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.3a1pre) Gecko/20100203 Minefield/3.7a1pre 2) set dom.ipc.plugins.enabled true 3) addons manager, add Flashblock 1.5.11.2 4) restart, and visit any youtube video site 5) Verify crash and a popup dialog saying: "A Plugin crashed while firefox was running..." (See screenshot) Expected: - no crash, and flashblock just works normally.
Blocks: OOPP
Whiteboard: [OOPPTestday]
No problem on Windows Vista so far.
Attached file console error messages (deleted) —
I see a similar problem on x86_64 Linux with Flashblock (1.5.11.2). I don't see a Crash Reporter dialog but on the console I see: Gdk-WARNING **: GdkWindow 0x3800003 unexpectedly destroyed The program '<unknown>' received an X Window System error.
I can reproduce with the nightlies of the past week or so on Linux.
Attachment #425118 - Attachment is obsolete: true
It does seem to give something in the reports now, at least for me: bp-b34970ab-610a-46a9-84e1-8b0792100213 0 libpthread-2.7.so libpthread-2.7.so@0x7540 1 libflashplayer.10.so libflashplayer.10.so@0x17c9c1 2 libflashplayer.10.so libflashplayer.10.so@0x1cac51 3 libflashplayer.10.so libflashplayer.10.so@0x49a5c 4 libflashplayer.10.so libflashplayer.10.so@0x3311f 5 libc-2.7.so libc-2.7.so@0x2e083 6 libgdk-x11-2.0.so.0.1200.9 libgdk-x11-2.0.so.0.1200.9@0x4c636 7 libX11.so.6.2.0 libX11.so.6.2.0@0x3873d 8 libX11.so.6.2.0 libX11.so.6.2.0@0x3fe5b 9 libX11.so.6.2.0 libX11.so.6.2.0@0x4071e 10 libX11.so.6.2.0 libX11.so.6.2.0@0x299d1 11 libgdk-x11-2.0.so.0.1200.9 libgdk-x11-2.0.so.0.1200.9@0x435ce 12 libgdk-x11-2.0.so.0.1200.9 libgdk-x11-2.0.so.0.1200.9@0x43a6e 13 libglib-2.0.so.0.1600.6 libglib-2.0.so.0.1600.6@0x37cf5 14 libglib-2.0.so.0.1600.6 libglib-2.0.so.0.1600.6@0x3b0b2 15 libglib-2.0.so.0.1600.6 libglib-2.0.so.0.1600.6@0x3b66d 16 libxul.so base::MessagePumpForUI::RunWithDispatcher ipc/chromium/src/base/message_pump_glib.cc:195 17 libxul.so base::MessagePumpForUI::Run ipc/chromium/src/base/message_pump_glib.h:59 18 libxul.so MessageLoop::RunInternal ipc/chromium/src/base/message_loop.cc:211 19 libxul.so MessageLoop::RunHandler ipc/chromium/src/base/message_loop.cc:194 20 libxul.so MessageLoop::Run ipc/chromium/src/base/message_loop.cc:168 21 libxul.so base::Thread::ThreadMain ipc/chromium/src/base/thread.cc:165 22 libxul.so _ZL10ThreadFuncPv ipc/chromium/src/base/platform_thread_posix.cc:26 23 libpthread-2.7.so libpthread-2.7.so@0x54fa 24 libc-2.7.so libc-2.7.so@0xd6e5d
Related to bug 545819 or bug 540114?
Interesting note: it crashes on view of the page, without clicking the placeholder, then crashes again on click of the placeholder. (same stack)
When you click on the placeholder the flash <object> is reinserted into the DOM so I'd expect the plugin to crash again.
I'd expect the crash to come on show, but is it possible to avoid letting it load enough to crash on initial page load? If this applies to all Flash, this of course means Flashblock is going to double Flash plugin crash report submission from all of its users.
We have a 10ms timeout before swapping out the flash object. We tried to remove this several times but without that timeout Firefox would crash frequently on pages with flash. You could edit our XPI and remove the timeout as an experiment.
Depends on: 540114
(In reply to comment #10) > We have a 10ms timeout before swapping out the flash object. All I find is this in flashblock.xml: var flashblockSetTimeout = nativeMethod(window, "setTimeout"); flashblockSetTimeout(flashblockShowPlaceholder, 0); Which is 0, not 10. This is in both 1.5.11.2 and 1.5.12. Anyway, if I comment that out and call flashblockShowPlaceholder() directly I don't get anything different. (and under 3.6 things still worked fine)
Assignee: nobody → karlt
Status: NEW → RESOLVED
Closed: 15 years ago
Resolution: --- → DUPLICATE
This seems to be fixed fully in bug 540114. The Youtube link in comment 4 crashes latest Minefield but in latest Electrolysis it all works fine. No crash on click or page load.
Status: RESOLVED → VERIFIED
Product: Core → Core Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: