dns_restart

Hello.
I am using Squid 2.2Stable5 with some H. Nordstrom patches in a Solaris
7 O.S.
I'm using dns_restart directive to restart dnsserver processes every
2000 requests.

# TAG: dns_restart
# dnsserver (or perhaps the resolver library) seems to be leaking
# memory. This option can be used to make the dnsserver slave
processes
# restart every now and then (restarts after N requests).
# The default is 0 to disable restarts.
#
#dns_restart 0
dns_restart 2000

When I look into cachemgr for the dnsserver statistics I always see a
#Request number higher than 2000 for every proccess (at least for the 4
or 5 firsts). If I look for the date the dnsserver proccess where
started (ps -aef |grep dnsserver) they all start around midnight (the
time I usualy rotate logs).
My question is who could I get sure the dns_restart directive is realy
working? I had supposed every dnsserver process might restart when it
reached more than 2000 requests solved.

Thanks in advance.

(P.S.: I beg your pardon for my bad English. I am not native english
speaker. )

-- 
-----------------------------------------------
Juan Carlos Sánchez Hernández				
Servicio de Planificación Informática			
Rectorado Universidad Politécnica de Madrid
Av. Ramiro de Maeztu, 7
Madrid - SPAIN
e-mail: jcsanchez@ccupm.upm.es
Tel.: +34 91 336 60 17
-----------------------------------------------