Additional info
linux:~ # uname -a
Linux linux 2.4.18-4GB #1 Wed Mar 27 13:57:05 UTC 2002 i686 unknown
linux:~ # dig
www.google.com
;; Total query time: 5012 msec
;; FROM: linux to SERVER: default -- 192.168.1.254
;; WHEN: Fri Jul 9 20:17:25 2004
;; MSG SIZE sent: 32 rcvd: 482
linux:~ # dig
www.google.com @10.0.0.6
;; Total query time: 16 msec
;; FROM: linux to SERVER: 10.0.0.6
;; WHEN: Fri Jul 9 20:24:22 2004
;; MSG SIZE sent: 32 rcvd: 482
linux:~ # dig
www.google.com @213.204.128.170
; <<>> DiG 8.3 <<>>
www.google.com @213.204.128.170
; (1 server found)
;; res options: init recurs defnam dnsrch
;; res_nsend to server 213.204.128.170: Connection timed out
Apparently, 10.0.0.6 is the working DNS (i guess the secondary one
might be down). Also, it seems the resolve.conf is not good enough
in the shape created by dhcpcd (not dhcpd, which I think I mistakenly
wrote previously). My guess is that the 5.012 seconds come from a 5
sec timeout when trying to use 192.168.1.254, which gives the same result as
213.204.128.170 (i.e. time out), plus the few milliseconds it takes to
get an answer from the 10.0.0.6.
But still, WHY?? This now seems as if the router puts its own addrerss
there in the responce to the dhcp request, and messes the resolve.conf up?
would it do that if it was a known fact that the result was like this?
Apparently to windows XP it does not matter. WHat is correct?
/jeff, increasingly puzzled, but stil lwanting to know why