Note: This is a public test instance of Red Hat Bugzilla. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback at bugzilla.redhat.com.
Bug 85060
Summary: | notification area segv | ||
---|---|---|---|
Product: | [Retired] Red Hat Linux | Reporter: | Matt Wilson <msw> |
Component: | gnome-panel | Assignee: | Havoc Pennington <hp> |
Status: | CLOSED RAWHIDE | QA Contact: | |
Severity: | medium | Docs Contact: | |
Priority: | medium | ||
Version: | 9 | CC: | otaylor |
Target Milestone: | --- | Keywords: | Triaged |
Target Release: | --- | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2003-10-21 20:11:17 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | |||
Bug Blocks: | 79578, 100643 |
Description
Matt Wilson
2003-02-25 06:47:24 UTC
<msw> hp: I was just sitting there <msw> hp: it came up 1 pixel wide <msw> hp: I right clicked on the 1 pixel in order to quit rhn <msw> hp: then I went and added the rhn applet again <msw> hp: and a little while later it crashed <msw> hah! <msw> duplicated! <msw> I had to log in N times to get it to be 1 pixel wide again <msw> right click on the notification icon, exit, then system tools->notification applet The situation seems to be a socket that's inside a plug that's inside a socket. Outermost socket activates the plug, which changes is_active, which then invokes socket_update_active(). So the innermost socket is in same process as the plug. The plug must be the applet and the socket is the system tray slot. Trying to see some reason socket_update_active wouldn't be removed while socket was invalid. Wild theory: hierarchy-changed is not received when destroying the system tray slot, because GTK_ANCHORED flag is not set on GtkPlug, and there's code in gtk_widget_unparent() that only does the hierarchy-changed if the toplevel for the window (in this case the plug) has GTK_ANCHORED set. If hierarchy-changed is not received when destroying the system tray slot socket, then socket_update_active() callback would not be removed and we would get a crash like this. That has to be crack though because it doesn't explain why it only happens sometimes; you'd expect it to crash any time you focused the applet after removing a tray icon. Probably 3am nonsense. In general, the ANCHORED flag is set on GtkPlug; it inherits from GtkWindow and GtkWindow sets it in it's _init(). There might be some problems if you reused a plug from in-process out-of-process. I'm not we update the flag properly when we make the GtkPlug a toplevel again, but that shouldn't matter here. Isn't this bug just the 1 pixel wide bug which has been fixed? yea, I've not been able to reproduce it anymore. |