Use "ps auxwwf" to figure out what are the zombies parents and kill the
parents. Whenever a process ends/dies/is killed it is turned into a
zombie until it is 'reaped' by it's parent process (the parent process
needs the zombie to stay around until it receives the SIGCHLD signal and
retrieves the process exit state). If a program has children but fails to
reap them (either because it doesn't handle SIGCHLD or reap in some other
manner) via lack of such facility or bad design or because it froze,
zombie processes may accumulate...
Cheers,
MaZe.
On Tue, 13 Dec 2005, Craig White wrote:
> after booting a CentOS 4.2 system
>
> I have a bunch of zombies...
>
> root 2828 0.0 0.0 0 0 ? Z 14:41 0:00 [chvt]
> <defunct>
>
> My read from googling this is chvt stems from something that involves a
> shell.
>
> They ***may*** have been created from my attempts to start mysqld which
> isn't working (I will start another thread) but no efforts to kill them
> off are working...
>
> [root@srv1 certs]# kill 2828
> [root@srv1 certs]# kill -9 2828
> [root@srv1 certs]# ps aux|grep 2828
> root 2828 0.0 0.0 0 0 ? Z 14:41 0:00 [chvt]
> <defunct>
>
> How do I clean this up?
>
> Craig
>
> _______________________________________________
> CentOS mailing list
> CentOS@xxxxxxxxxx
> http://lists.centos.org/mailman/listinfo/centos
>
