Closed Bug 732004 Opened 13 years ago Closed 8 years ago

Hang with 180% cpu on mac

Categories

(Core Graveyard :: Plug-ins, defect)

x86
macOS
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED INCOMPLETE

People

(Reporter: Fallen, Unassigned)

Details

Attachments

(5 files)

I clicked on a xing.com link from within Thunderbird, which caused Firefox to open it. It got stuck in the "connecting..." state, so I left it there and continued work. After a while I noticed that I had high CPU, OSX ps says ~180%. I'm not sure if its the xing page that causes the problems, but this is what I remember. The exact link was: http://www.xing.com/go/contacts?reagent=systemmail/mailcontactaddnotify1&dpt=sysm&ttp=mail&plf=own&pid=myctc&xtr=lnk I've done a few samples using the OSX activity monitor, I hope it helps. 2/3 Samples contain: 1 mac_plugin_interposing_child_OnSetCursor (in XUL) + 757358 [0x101ec58ce] 1 XRE_AddStaticComponent (in XUL) + 21393 [0x101f04c31]
Seems the flash and Google Talk plugin were also active. Killing these two didn't make Firefox come back kewisch 13121 0.0 0.0 702704 1056 ?? S 23Feb12 0:02.31 /Applications/Firefox.app/Contents/MacOS/plugin-container.app/Contents/MacOS/plugin-container /Library/Internet Plug-Ins/googletalkbrowserplugin.plugin -greomni /Applications/Firefox.app/Contents/MacOS/omni.ja 397 gecko-crash-server-pipe.397 org.mozilla.machname.61567877 plugin kewisch 401 0.0 0.6 2893012 25200 ?? S 17Feb12 145:44.20 /Applications/Firefox.app/Contents/MacOS/plugin-container.app/Contents/MacOS/plugin-container /Library/Internet Plug-Ins/Flash Player.plugin -greomni /Applications/Firefox.app/Contents/MacOS/omni.ja 397 gecko-crash-server-pipe.397 org.mozilla.machname.1600164120 plugin
It would probably be useful to attach here the contents of about:support and about:plugins. Also, (sorry I forgot to ask for that), during the hang, you could also sample the plugin_container process.
Attached file about:plugins
Attached file about:support
Unfortunately I've killed the process by now, so I can't sample the plugin container.
Also let us know exactly which versions of Firefox and Thunderbird you can reproduce this with. And tell us if you *can* reproduce it. You appear to be running on OS X 10.7.3. I suspect many of the symbols in your traces are spurious ... though not all of them are. (This is because Firefox distros have at least some of their symbols stripped.)
Unfortunately not reproduceable. Also, firefox updated just after the crash, so it may be inaccurate. I've been using Firefox 11 betas, so I suspect its the next-to-latest beta. I doubt Thunderbird has to do with this since it probably opened it via external URL service. Anyway, its also Thunderbird 11 latest beta.
(In reply to comment #3) Just noticed something rather odd about these two plugin-container processes: The Google Talk plugin process had been running for 2.5 minutes, but the Flash plugin process had been running for several days. Was Firefox already running when you triggered the hang (by clicking that link in Thunderbird)? Were multiple copies of Firefox running?
> The Google Talk plugin process had been running for 2.5 minutes Actually, if the Google Talk plugin process started on 23Feb12, it had also been running for several days. But it had only used 2.5 minutes "of CPU time".
Attachment #601973 - Attachment mime type: text/html → application/xhtml+xml
Attachment #601973 - Attachment mime type: application/xhtml+xml → text/html
Firefox was already running, I closed/opened the lid multiple times in between.
Resolving old bugs which are likely not relevant any more, since NPAPI plugins are deprecated.
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → INCOMPLETE
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: