I have an existing NW6 server operating as the sole DNS server for a long
time that is now to be replaced in this role by a new NW65 server.

Having configured this new server as a DNS server, loaded named.nlm on it
and changed DHCP options for my workstation so that I alone point to the new
server (for testing) I find that every initial Nslookup request I send fails
with a time-out, as per the following (debug) message:

> www.bmw.com

Server: news.intranet <----- second alias in DNS of new DNS server
Address: XXX.XXX.XXX.42 <----- new DNS server IP address

------------
Got answer:
HEADER:
opcode = QUERY, id = 9, rcode = NXDOMAIN
header flags: response, auth. answer, want recursion, recursion
avail.
questions = 1, answers = 0, authority records = 1, additional =
0

QUESTIONS:
www.bmw.com.intranet, type = A, class = IN
AUTHORITY RECORDS:
-> intranet
ttl = 86400 (1 day)
primary name server = DNSNAME.intranet <---- old/current DNS
server
responsible mail addr = root.intranet
serial = 2002062013
refresh = 10800 (3 hours)
retry = 3600 (1 hour)
expire = 604800 (7 days)
default TTL = 86400 (1 day)

------------
DNS request timed out.
timeout was 2 seconds.
timeout (2 secs)
*** Request to news.intranet timed-out

(where news.intranet is one of the DNS aliases of the new server - not the
actual server name though, strangely so do not know why it randomly picks
the non-primary name of the server for some of these nslookup requests)


BUT then a *repeat* of the same request immediately after produces an answer
!


> www.bmw.com

Server: news.intranet
Address: XXX.XXX.XXX.42

------------
Got answer:
HEADER:
opcode = QUERY, id = 11, rcode = NXDOMAIN
header flags: response, auth. answer, want recursion, recursion
avail.
questions = 1, answers = 0, authority records = 1, additional =
0

QUESTIONS:
www.bmw.com.intranet, type = A, class = IN
AUTHORITY RECORDS:
-> intranet
ttl = 86400 (1 day)
primary name server = DNSNAME.intranet
responsible mail addr = root.intranet
serial = 2002062013
refresh = 10800 (3 hours)
retry = 3600 (1 hour)
expire = 604800 (7 days)
default TTL = 86400 (1 day)

------------
------------
Got answer:
HEADER:
opcode = QUERY, id = 12, rcode = NOERROR
header flags: response, want recursion, recursion avail.
questions = 1, answers = 5, authority records = 9, additional =
0

QUESTIONS:
www.bmw.com, type = A, class = IN
ANSWERS:
-> www.bmw.com
canonical name = bmw.com.edgesuite.net
ttl = 3565 (59 mins 25 secs)
-> bmw.com.edgesuite.net
canonical name = bmw.georedirector.akadns.net
ttl = 21566 (5 hours 59 mins 26 secs)
-> bmw.georedirector.akadns.net
canonical name = a250.g.akamai.net
ttl = 267 (4 mins 27 secs)
-> a250.g.akamai.net
internet address = 144.135.8.151
ttl = 20 (20 secs)
-> a250.g.akamai.net
internet address = 144.135.8.207
ttl = 20 (20 secs)
AUTHORITY RECORDS:
-> g.akamai.net
nameserver = n5g.akamai.net
ttl = 1768 (29 mins 28 secs)
-> g.akamai.net
nameserver = n6g.akamai.net
ttl = 1768 (29 mins 28 secs)
-> g.akamai.net
nameserver = n7g.akamai.net
ttl = 1768 (29 mins 28 secs)
-> g.akamai.net
nameserver = n8g.akamai.net
ttl = 1768 (29 mins 28 secs)
-> g.akamai.net
nameserver = n0g.akamai.net
ttl = 1768 (29 mins 28 secs)
-> g.akamai.net
nameserver = n1g.akamai.net
ttl = 1768 (29 mins 28 secs)
-> g.akamai.net
nameserver = n2g.akamai.net
ttl = 1768 (29 mins 28 secs)
-> g.akamai.net
nameserver = n3g.akamai.net
ttl = 1768 (29 mins 28 secs)
-> g.akamai.net
nameserver = n4g.akamai.net
ttl = 1768 (29 mins 28 secs)

------------
Non-authoritative answer:
Name: a250.g.akamai.net
Addresses: 144.135.8.151, 144.135.8.207
Aliases: www.bmw.com, bmw.com.edgesuite.net
bmw.georedirector.akadns.net


I assume from here on the result is cached locally and so that is why it
continues to succeed for the same domain..

Now, why does the current NW6 DNS server always succeed first time and the
new NW65 server always fail first time then succeed subsequently?

Both servers have eDirectory replicas locally and I setup the Forwarding and
Zone info identically in both server properties in the DNS/DHCP Management
Console so there shouldn't be greater delays for one server than the other
in resolving a non-cached address..