After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 122263 - Eats lot of CPU together with the acme daemon
Eats lot of CPU together with the acme daemon
Status: RESOLVED DUPLICATE of bug 122327
Product: gnome-panel
Classification: Other
Component: notification area
2.2.x
Other Linux
: Normal normal
: ---
Assigned To: Havoc Pennington
Panel Maintainers
Depends on:
Blocks:
 
 
Reported: 2003-09-14 09:30 UTC by Jean-Luc Coulon
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Jean-Luc Coulon 2003-09-14 09:30:38 UTC
Hi,
Together, the notification area applet and the acme daemon are eating 50%
of the cpu on a K6-2 500MHz.

Removing the applet suppresses the problem, even if acme is running.

If I suppress the applet and kill acme then reinstall 1st the applet then
launch acme, all went fine.

If I restart gnome, then I can see the same high cpu load as before.

---
Regards
        Jean-Luc
Comment 1 Jean-Luc Coulon 2003-09-17 06:27:18 UTC
Hi,

I've done some strace on the running pid.

1 - I start galeon/metacity from gdm with the notification area applet
and acme running (attached to the applet).

2 - I kill acme and remove the applet from the gnome-panel.

I left the trace running a couple of seconds and I got the follwing files:

-rw-r--r--    1 jean-luc jean-luc   884738 2003-09-16 17:44 trace.not.1
-rw-r--r--    1 jean-luc jean-luc      627 2003-09-16 18:25 trace.not.2

If then I close the session and I restart galeon, I'm back in the case
one.

The traces now:


Most of the trace.not.1 file contains :

Process 9973 attached - interrupt to quit
...
read(3, "\1\1\201i\0\0\0\0A\0\0\3\0\0\0\0\0\0\0\0\36\0\0\0xe\221"...,
32) = 32
write(3, "\31\0\v\0\34\0 \2\0\0\0\0! \7\10\34\0 \2\200\1\0\0\0\0"...,
48) = 48
read(3, "\1\1\203i\0\0\0\0A\0\0\3\0\0\0\0\0\0\0\0\36\0\0\0xe\221"...,
32) = 32
write(3, "\31\0\v\0\34\0 \2\0\0\0\0! \7\10\34\0 \2\200\1\0\0\0\0"...,
48) = 48
read(3, "\241 \203i\37\0\300\2\200\1\0\0\0\0\0\0\6\0\0\0\0\0\0\0"...,
32) = 32
read(3, "\1\1\205i\0\0\0\0A\0\0\3\0\0\0\0\0\0\0\0\36\0\0\0xe\221"...,
32) = 32
write(3, "\31\0\v\0\236\0 \1\0\0\0\0! \7\10\236\0 \1\200\1\0\0\0"...,
48) = 48

.... add many ...

write(3, "\31\0\v\0\236\0 \1\0\0\0\0! \7\10\236\0 \1\200\1\0\0\0"...,
48) = 48
read(3, "\1\1\215i\0\0\0\0A\0\0\3\0\0\0\0\0\0\0\0\36\0\0\0xe\221"...,
32) = 32
write(3, "\31\0\v\0\34\0 \2\0\0\0\0! \7\10\34\0 \2\200\1\0\0\0\0"...,
48) = 48
read(3, "\1\1\217i\0\0\0\0A\0\0\3\0\0\0\0\0\0\0\0\36\0\0\0xe\221"...,
32) = 32

With some :
ioctl(3, FIONREAD, [0])                 = 0
ioctl(3, FIONREAD, [0])                 = 0
poll([{fd=10, events=POLLIN}, {fd=3, events=POLLIN, revents=POLLIN},
{fd=15, 
eve
nts=POLLIN}, {fd=17, events=POLLIN|POLLPRI}, {fd=18,
events=POLLIN|POLLPRI}, 
{fd
=19, events=POLLIN|POLLPRI}, {fd=21, events=POLLIN|POLLPRI}, {fd=20, 
events=POLL
IN|POLLPRI}, {fd=22, events=POLLIN|POLLPRI}, {fd=23,
events=POLLIN|POLLPRI}], 
10
, -1) = 1
ioctl(3, FIONREAD, [32])                = 0

and some :

read(3, 0xbfffd5f0, 32)                 = -1 EAGAIN (Resource temporarily 
unavailable)



The whole trace.not.2 contains:

Process 2482 attached - interrupt to quit
octl(3, FIONREAD, [0])                 = 0
poll([{fd=10, events=POLLIN}, {fd=3, events=POLLIN, revents=POLLIN},
{fd=15, 
eve
nts=POLLIN}, {fd=17, events=POLLIN|POLLPRI}, {fd=18,
events=POLLIN|POLLPRI}, 
{fd
=19, events=POLLIN|POLLPRI}, {fd=21, events=POLLIN|POLLPRI}, {fd=20, 
events=POLL
IN|POLLPRI}, {fd=22, events=POLLIN|POLLPRI}, {fd=23,
events=POLLIN|POLLPRI}], 
10
, -1) = 1
ioctl(3, FIONREAD, [32])                = 0
read(3, "n\2_\1\234\21\16\0\4\24\4\0\20\0\0\0\0\0\0\24\24\24\24"...,
32) = 32
ioctl(3, FIONREAD, [0])                 = 0
poll( <unfinished ...>
Process 2482 detached

---
Best Regards
Comment 2 Herbie Hopkins 2003-09-18 10:14:10 UTC
I'm experiencing the same problems with gnome-panel 2.4.0 only it's
eating 100% cpu on my athlon-xp 2100. I have also found that it's not
purely due to acme since I experience the same thing with gaim's
notification area plugin. It only seems to occur when the notification
applet is started after either gaim or acme and I have also noticed
that the cpu usage drops to normal if I left click on the notification
area.

Herbie.
Comment 3 Christian Marillat 2003-09-24 12:45:33 UTC
Bug fixed in debian, see latest post to http://bugs.debian.org/210813
Comment 4 Jean-Luc Coulon 2003-09-25 17:03:55 UTC
Hi,

I've reported the bug (I think)
Nothing has be done here or in Debian.
I would like you reopen the bug please as it is already existing on my
system (up-ro-date regarding Debian sid packages).

---
Regards
        Jean-Luc
Comment 5 Frederic Crozat 2003-10-03 09:26:02 UTC

*** This bug has been marked as a duplicate of 122327 ***