[CentOS] Bind on Centos4

I did that ipv6 change and it has made a significant difference (400 ms 
instead of 4000) -

[root at host3 ~]# dig cisco.com

; <<>> DiG 9.2.4 <<>> cisco.com
;; global options:  printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 59188
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 2, ADDITIONAL: 0

;; QUESTION SECTION:
;cisco.com.                     IN      A

;; ANSWER SECTION:
cisco.com.              86400   IN      A       198.133.219.25

;; AUTHORITY SECTION:
cisco.com.              86400   IN      NS      ns1.cisco.com.
cisco.com.              86400   IN      NS      ns2.cisco.com.

;; Query time: 415 msec
;; SERVER: 127.0.0.1#53(127.0.0.1)
;; WHEN: Thu May 12 10:20:34 2005
;; MSG SIZE  rcvd: 79


Troy Engel wrote:
> Someone else recently had this problem (search the archives a few days 
> back). His solution was to rebuild a SRPM from Fedora Core of an older 
> BIND and it solved it.
> 
> I suggested it might be IPv6 lookups stalling you; try adding:
> 
>   alias net-pf-10 off
> 
> to your /etc/modprobe.conf and reboot the server, see if that fixes the 
> slowness issue. This used to show up when IPv6 was first introduced into 
> Fedora Core, Mozilla had lookup pause issues as well.
> 
> -te
> 
> Tony Wicks wrote:
> 
>> identical (both have caching-nameserver removed). The Centos 4 machine 
>> is very slow at initial lookups -
> 
>