Description of problem: esd runaway process
Version-Release number of selected component (if applicable):
How reproducible: Every reboot
Steps to Reproduce:
1. Reboot into 2.6.21-rc3
2. All is well
3. after an hour of work: email (thunderbird), web browsing (firefox), etc.
Snapshot of w:
[root@mozart log]# w
20:25:18 up 2:34, 1 user, load average: 2.74, 2.43, 2.21
USER TTY FROM LOGIN@ IDLE JCPU PCPU WHAT
kaune pts/1 :0.0 17:53 0.00s 0.04s 1.51s gnome-terminal
nothing else is going on except this web browser entering the bugzilla data.
Snapshot of top:
top - 20:21:43 up 2:30, 1 user, load average: 2.21, 2.17, 2.10
Tasks: 140 total, 4 running, 135 sleeping, 0 stopped, 1 zombie
Cpu(s): 17.6%us, 82.4%sy, 0.0%ni, 0.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 1036268k total, 749316k used, 286952k free, 48760k buffers
Swap: 2015992k total, 0k used, 2015992k free, 515148k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3687 kaune 25 0 3716 1688 980 R 98.5 0.2 142:23.29 esd
3700 kaune 15 0 39464 14m 10m R 0.7 1.4 0:01.32 gnome-terminal
3574 root 15 0 107m 20m 7756 S 0.3 2.0 1:12.17 Xorg
3663 kaune 15 0 12664 4132 3264 S 0.3 0.4 0:17.76 at-spi-registry
4622 root 15 0 2192 1008 772 R 0.3 0.1 0:00.02 top
1 root 15 0 2128 632 544 S 0.0 0.1 0:00.47 init
esd is running full tilt and no real warning. I have looked in
/var/log/messages and do not know where else to get a clue as to why it has run
Obviously, it is not expected to be the cause of a 100% cpu utilization.
I am fully up to date as far as yum update provides at this writing. This has
been occuring for at least the last week. My solution is to kill -TERM the
pid- but ...
My system is the Fedora development on a P4 2.4 GHz with 1Gb RAM. I am not
running the distributed kernel, but rather a "pure" 2.6.21-rc3 kernel, because
so far, I have issues with my hda, hdb and sata_via sda, sdb and scsi drive sdc
being found properly. (all are found as sd? and not as my setup expects, hence
the boot panics) Naturally, this is a whole other issue I want to figure out
and will file under a separate bug).
It appears the new alsa updates may have resolved whatever issue this was. The
problem has disappeared following the update this morning. Thank you
Kirk C Aune