Thanks to all who replied. I decided to wait out the 48hrs left on the local nameserver
(Hugo's tip) and then the problem sorted itself. :)
Valuable tips might like these ought to be compiled into a sysadmin pocketbook:
'wisdom from bitfolk'.
Cheers!
--- On Tue, 2/7/12, Hugo Mills <hugo-bf(a)carfax.org.uk> wrote:
From: Hugo Mills <hugo-bf(a)carfax.org.uk>
Subject: Re: [bitfolk] Renumbering: resolver not working? still points to 212.x, instead
of 85.y on >=1 machine
To: "Max B" <txtmax(a)yahoo.ca>
Cc: users(a)lists.bitfolk.com
Received: Tuesday, February 7, 2012, 9:59 PM
On Tue, Feb 07, 2012 at 12:56:19PM -0800, Max B wrote:
thanks for everyone's response to my wiki
proposal. I hadn't
intended a flame war, and may respond selectively at a future date,
when for instance, I am not faced with a puzzle, as follows.
dig returns the correct value 85.y on my VPS.
dig returns the old address 212.x elsewhere (at work).
Two things you can look at: dig tells you the TTL remaining on any
entry when you query it. For example:
hrm@ruthven:~ $ dig oxfordflutes.co.uk
; <<>> DiG 9.7.3 <<>> oxfordflutes.co.uk
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 52296
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 0
;; QUESTION SECTION:
;oxfordflutes.co.uk. IN A
;; ANSWER SECTION:
oxfordflutes.co.uk. 86397 IN A 85.119.82.48
;; Query time: 0 msec
;; SERVER: 10.0.0.10#53(10.0.0.10)
;; WHEN: Tue Feb 7 21:50:55 2012
;; MSG SIZE rcvd: 52
The "86397" in the ANSWER SECTION is the TTL, and shows that I've
got a few seconds short of 24 hours left for this entry in the cache.
This value will count down to zero, at which point the cache expires
and a new authoritative entry (hopefully the right one) will be
retrieved.
The other thing is to use "@servername" in your dig query, which
will explicitly bypass the locally configured resolvers, and query the
named DNS server.
hrm@ruthven:~ $ dig @ns1.bb-online.net. oxfordflutes.co.uk
[...]
In this case, I'm querying one of the definitive servers for
oxfordflutes.net, and so should get the results that I've asked them
to publish. You can check whether the DNS servers under your control
(usually the primary and one or more secondaries) are sending the
right answers. If they are, then you should simply have to wait until
the TTL in the previous query reaches zero. If not, then you have some
kind of misconfiguration of your servers.
Oh, and you should be able to get the definitive list of
nameservers by issuing an NS query:
hrm@ruthven:~ $ dig NS oxfordflutes.co.uk
; <<>> DiG 9.7.3 <<>> NS oxfordflutes.co.uk
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 43912
;; flags: qr rd ra; QUERY: 1, ANSWER: 3, AUTHORITY: 0, ADDITIONAL: 0
;; QUESTION SECTION:
;oxfordflutes.co.uk. IN NS
;; ANSWER SECTION:
oxfordflutes.co.uk. 86400 IN NS brain.bb-online.co.uk.
oxfordflutes.co.uk. 86400 IN NS
ns1.bb-online.net.
oxfordflutes.co.uk. 86400 IN NS
ns2.bb-online.net.
;; Query time: 48 msec
;; SERVER: 10.0.0.10#53(10.0.0.10)
;; WHEN: Tue Feb 7 21:54:55 2012
;; MSG SIZE rcvd: 115
So here my definitive nameservers are brain, ns1 and
ns2.bb-online.net. Check the data returned by all of the servers
(with the @server syntax above) for the NS query, just to see if one
is out of sync with the others.
Hugo.
--
=== Hugo Mills: hugo@... carfax.org.uk |
darksatanic.net | lug.org.uk ===
PGP key: 515C238D from
wwwkeys.eu.pgp.net or
http://www.carfax.org.uk
--- UNIX: Japanese brand of food containers. ---