named gets swapped in and out

[phn at icke-reklam.ipsec.nu]

Marc Haber <mh+usenet0236 at zugschl.us> wrote:

> phn at icke-reklam.ipsec.nu wrote:
>>Ignacio <engineering at cableonda.net> wrote:
>>> What could be the problem?  It could be hardware or some wrong bind
>>> configuration?  Shuld I change max recursive-clients, max-cache-size
>>> or something like this?
>>
>>The problem is that you are lacking memory. Buy more. Memory=20
>>needs will grow after start until a balance between questions=20
>>asked by clients and the lifetime of cached data.

> Pardon me for jumping in, but I experience the same problem. We have
> two identical machines, configured identically. They have the same
> software, they run the same version of BIND (9.2.1 on Debian woody),
> and they load the same master and slave zones. They don't answer any
> recursive queries, so their caching behavior should be about the same
> as well.

> One of the machines frequently has its named eat over a hundred
> Megabytes, forcing it into swap, while the other one is chugging away
> with its named process at about 40 MB.

> I don't understand why one of the boxes behaves so differently than
> the other one. Any ideas?

if query-logging is considered to heavy, a dump with rndc dumpdb 
will give clues of what teh cache consist of.


> Greetings
> Marc

> -- 
> -------------------------------------- !! No courtesy copies, please !! -----
> Marc Haber          |   " Questions are the         | Mailadresse im Header
> Karlsruhe, Germany  |     Beginning of Wisdom "     | Fon: *49 721 966 32 15
> Nordisch by Nature  | Lt. Worf, TNG "Rightful Heir" | Fax: *49 721 966 31 29


-- 
Peter Håkanson         
        IPSec  Sverige      ( At Gothenburg Riverside )
           Sorry about my e-mail address, but i'm trying to keep spam out,
	   remove "icke-reklam" if you feel for mailing me. Thanx.