Urgent: Mad Bind

[Birgit Schmid]

On Fri, Jan 28, 2000 at 08:51:45AM +0000, Jim Reid wrote:
> >>>>> "Birgit" == Birgit Schmid <Birgit.Schmid at ecrc.de> writes:
> 
>     Birgit> my bind starts to "consume" a little bit too much:
> 
>     Birgit> CPU states: 0.0% idle, 89.0% user, 10.8% kernel, 0.2% iowait, 0.0% swap
>     Birgit> Memory: 512M real, 8112K free, 58M swap in use, 2387M swap free
> 
>     Birgit>  PID USERNAME THR PRI NICE  SIZE   RES STATE   TIME    CPU COMMAND
>     Birgit>  906 root       1   0    0   49M   39M run    50:35 98.23% named
> 
> What version of the OS and BIND are you running?

Solaris 2.6 and Bind8.2.2P5.

> 
> It might also be an idea to use a system call tracer like truss or
> ktrace to find out what the process is doing. The info you've given

It was complaining about a too big database file. unfortunately i lost
the exact error.

> 
> Could it be that you've just caught the name server while it was in
> the middle of cleaning up its cache? That could account for a lot of
> user-mode CPU activity for a few seconds every hour or so.

is there anything to state this theory?

I found some files named xfer.ddt.*. after deleting them and a restart
bind feels a lot better.

12845 root       1  -2    0   60M   57M run   113:59 14.55% named

Can anyone enlighten me on this?

Thanks a lot.

-birgit