Re: BIND process size

[Alexey Fadyushin <fab s-tunnel com>]

Because DNS queries ahould be answered very quickly, ndmed tries to 
cache as many DNS records it learns from other servers as possible (and 
as limited by TTL fields in those records). So, the named daemon could 
stabilize using tens of megabytes of RAM. For example, on my servers 
after running for long time (months), it uses from 14MB to 37MB, 
depending on the number of zones it holds and on intensity of requests 
from its clients.

If you want to limit the amount of memory used by named, consider using 
the option 'datasize' in named.conf.

Alexey Fadyushin
Brainbench MVP for Linux.
http://www.brainbench.com

Rushan Sobar wrote:

> Hi,
>
> am running RedHat 9 as main DNS server with bind-9.2.1-16
> when I run top command on the uptime around 10 days I can see
>  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME CPU COMMAND
> 28574 named     25   0 25165  24M   988 S     0.5  4.7     2:33      0   named
>
> the process size 24M do I have to set a cron job to restart the named every specific time so I refresh and decrease the size of the named
> or its ok and will work fine even if the machine run for long time ?
>
> Advise...
>
> Regards
> Rushan