Bug 316521

Summary: clock applet quits repeatedly
Product: [Fedora] Fedora Reporter: Michael V. Antosha <mivael>
Component: gnome-panelAssignee: Ray Strode [halfline] <rstrode>
Status: CLOSED WONTFIX QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: low Docs Contact:
Priority: low    
Version: 6   
Target Milestone: ---   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2008-03-18 18:40:50 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:
Attachments:
Description Flags
clock applet died, backtrace (X Window System Error, BadRequest) none

Description Michael V. Antosha 2007-10-03 07:57:13 UTC 
Description of problem:

Please, refer to Bug 277941 for detailed description of the problem.

In short: my clock applet quits from time to time, without any obvious reason.
I suspect that the original cause might be the same as in bug 282691 (xchat),
bug 301051 (thunderbird), bug 308701 (firefox).

Now I'm going to follow the advice which Ray Strode gave me recently (Bug
277941, Comment #17), i.e. to run the clock-applet from under GDB.


Version-Release number of selected component (if applicable):

   gnome-panel-2.16.3-2.fc6.src.rpm


Steps to Reproduce:
1. Use clock applet for a long time (several days or weeks), without exiting X
Window session
2. Wait for dialog window that says that "Clock" has quit unexpectedly.
  
Actual results:

Clock applet disappears. Dialog window appears that reports about the problem
and suggests to "Reload" the applet. After pressing the "Reload" button, the
clock applet appears again and works properly, till the next same problem.
Comment 1 Michael V. Antosha 2007-10-03 09:22:50 UTC 
Ray, I couldn't follow your advice exactly.
The sequence I used was like that:

1. killing clock-applet process
2. waiting for the dialog window and pressing "Reload" button
3. ensuring that clock-applet is running (PID of the process needed)
4. making gdb to attach to the process with that PID:

   gdb /usr/libexec/clock-applet 1811   # 1811 is a PID of the clock applet

(process stopped now, by gdb)

5. resuming the process:

   (gdb) set height 0        # pagination...
   (gdb) break gdk_x_error   # just in case the problem is the same
   (gdb) continue

Now the clock applet works under GDB, I'm waiting for the next crashing.
Comment 2 Matthias Clasen 2007-10-03 21:53:55 UTC 
To make this really useful, you want to run the clock applet with --sync
Comment 3 Michael V. Antosha 2007-10-04 07:16:00 UTC 
Oh, I forgot... sorry  :(

Well... but I don't really know how to do it right.

The applet, I guess, needs special environment settings and, maybe, more tricks
to make it work under GDB? Please, give me a hint about how to do this right.
Comment 4 Michael V. Antosha 2007-10-10 06:51:50 UTC 
Created attachment 222221 [details]
clock applet died, backtrace (X Window System Error, BadRequest)

The problem proved to be similar to those described in the bug reports
mentioned above.

Details:

I run the clock-applet under gdb with the same parameters as displayed by "ps
axww | fgrep clock" (when the applet started in usual way), and one more option
("--sync"). I set a breakpoint on gdk_x_error and "run" the process. Then, I
reload the clock applet (either with the help of dialog window, or with the
"Add to panel..." feature. In both cases no additional processes appear, clock
applet works fine (until it gets to the breakpoint).

The attachment contains a stack trace of the clock-applet caught at
gdk_x_error(). After I "continue" running the process, it dies with the same
(BadRequest) X Window error as in related bug reports.
Comment 5 Michael V. Antosha 2007-10-10 07:17:40 UTC 
Meanwhile... Bug 318561 is my trying to tie all those similar problems and to
resolve them from the other side (libX11).
Comment 6 Ray Strode [halfline] 2008-03-18 18:40:50 UTC 
Hi,

We no longer support Fedora Core 6 and I am currently trying to get my open bug
count down to a more manageable state.  I'm going to close this bug as WONTFIX.
 If this issue is still a concern for you, would you mind trying to reproduce on
a supported version of Fedora and reopening?

(this is a mass message)