named cause defunct processes
Mark.Andrews at nominum.com
Mark.Andrews at nominum.com
Fri Sep 29 23:34:14 UTC 2000
Pick up BIND 8.2.3 BETA. There is internal scheduling problem
in BIND 8.2.2-P5 and some earlier versions that cause named not
to reap the child processes under high query loads. The children
will get reaped when the query load drops.
Mark
>
>
> Hi,
>
> recently i have a problem with my dns that cause it to work very slow.
> named daemon "eats" CPU "95.90% named" and when i check the processes
> that are running i found out lot of defunct processes , i checked parent
> process of the defunct process and it's the named process.
>
> ps auxw | grep defunct
> root 4963 0.0 0.0 0 0 Z 0:00 <defunct>
> root 4982 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5005 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5020 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5035 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5040 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5042 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5043 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5044 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5045 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5046 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5048 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5050 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5053 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5055 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5070 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5077 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5079 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5083 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5103 0.0 0.0 0 0 Z 0:00 <defunct>
> root 5157 0.0 0.0 0 0 Z 0:00 <defunct>
>
>
>
> I am running Solaris 2.8 32bit ,2GB RAM on sun E220 server, named 8.2.2-P5
>
> ps: i replaced the current server with other server that runs solaris 2.7
> same problem.
>
>
> -rabia.
>
>
>
>
>
--
Mark Andrews, Nominum Inc.
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: Mark.Andrews at nominum.com
More information about the bind-users
mailing list