dig/nslookup works remotly, but not locally (using our nameservers both times)....., Strange DNS problem Options

[Laeelin]

To proect the innocent, I'll be using the names below:
Nameserver: ns1.site.com
Our Web server: server.site.com
Our Web Site: somesite.com (we have many sites, all have the same problem)
All IP addresses: XXX.XXX.XXX.??? (replacing ??? with the correct information)

Now right into the wierd part...

From my work computer:

[/]# nslookup somesite.com ns1.site.com
Server: ns1.site.com
Address: XXX.XXX.XXX.194#53

Name: somesite.com
Address: XXX.XXX.XXX.202

exact same command from our web server...

From the web server (using ssh)

[/]# nslookup somesite.com ns1.site.com
Server: ns1.site.com
Address: XXX.XXX.XXX.194#53

** server can't find somesite.com: SERVFAIL

However other sites (like google) work.

From the web server (using ssh)

[/]# nslookup google.com ns1.site.com
Server: ns1.site.com
Address: XXX.XXX.XXX.194#53

Non-authoritative answer:
Name: google.com
Address: 64.233.167.99
Name: google.com
Address: 64.233.187.99
Name: google.com
Address: 72.14.207.99

We have a new server with running RedHat, WHM and cpanel.

WHM 11.11.0 cPanel 11.16.0-S18450
REDHAT Enterprise 5.1 i686 on standard - WHM X v3.1.0

When I dig, I get this:

[/]# dig @ns1.site.com somesite.com

; <<>> DiG 9.2.4 <<>> @ns1.site.com somesite.com
; (1 server found)
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 788
;; flags: qr aa rd; QUERY: 1, ANSWER: 1, AUTHORITY: 2, ADDITIONAL: 2

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

;; ANSWER SECTION:
somesite.com. 14400 IN A XXX.XXX.XXX.202

;; AUTHORITY SECTION:
somesite.com. 86400 IN NS ns2.site.com.
somesite.com. 86400 IN NS ns1.site.com.

;; ADDITIONAL SECTION:
ns1.site.com. 14400 IN A XXX.XXX.XXX.194
ns2.site.com. 14400 IN A XXX.XXX.XXX.194

;; Query time: 10 msec
;; SERVER: XXX.XXX.XXX.194#53(XXX.XXX.XXX.194)
;; WHEN: Sun Jan 13 16:36:34 2008
;; MSG SIZE rcvd: 125

From the web server...

[/]# dig @ns1.site.com somesite.com

; <<>> DiG 9.3.3rc2 <<>> @ns1.site.com somesite.com
; (1 server found)
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 31172
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 0

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

;; Query time: 14 msec
;; SERVER: XXX.XXX.XXX.194#53(XXX.XXX.XXX.194)
;; WHEN: Sun Jan 13 15:40:12 2008
;; MSG SIZE rcvd: 33

If you want any other information, please ask and I'll post tomorrow,

[filmo]

is your ns1.site.com

name server by chance one run by EV1 or The Planet?

We're also having very strange and sporadic connectivity issues with our scripts the need to resolve DNS/IP addresses.

In particular, they are failing during 'peak' business hours but run fine during off-hours (after 5:00 PM PST).

For example PHP scripts that connect to the mail server give random but frequent failure to connect errors during business hours, but run perfectly fine after hours.

Like wise, we maintain two sites on the same server that make SOAP calls to each other.

www.my_first_site.com makes a SOAP call to www.my_second_site.com.

Part of the SOAP process for the calling script is to resolve the IP of the second site (even though they are on the same server). Again, during normal business hours, slow performance or failure. After hours, speed performance consistently.

We do not host DNS on our server.

I'm starting to think that perhaps is the nameserver is set to something controlled by The Planet as a shared resource, it over taxed or over subscribed.