Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Simple setup:

  • Primary server (\MASTER): Windows 2008 server standard with following roles

    • DHCP server
    • DNS server
    • AD domain controller
  • Secondary server (\SLAVE): Windows 2003 server standard with following roles

    • DNS server
    • AD domain controller (additional)

LAN uses private IPs of range 10.10.1.x. No subnets. Primary server (MASTER) is configured to accept incoming VPN connections and give them IP addresses from 10.10.1.192~240 range.

Everything is working fine. AD and DNS replication is ok. Remote clients can connect into network and access it.

However, sometimes the DNS resolution for MASTER returns 10.10.1.192 (or .193, .194, etc). I go to DNS database and I discover that it contains TWO entries: one for MASTER as 10.10.1.1 (correct) and ANOTHER for MASTER as 10.10.1.192 (incorrect).

This, as you can suppose, causes us a big headache since the machines on LAN suddenly stop finding MASTER server by its DNS name. I had to remove manually the incorrect (A) entry from DNS database and run a IPCONFIG /FLUSHDNS in all workstations to fix this.

So, the question is: WHO are creating the new (incorrect) entries into the DNS Server database? Why? And HOW I disable/fix it?


UPDATE

To the ones whom I could not made myself entirely clear WHY there IS a problem:

Problem:

C:\>ping MASTER

Pinging 10.10.1.193 with 32 bytes of data:

Request timed out.
Request timed out.
Request timed out.
Request timed out.

Ping statistics for 10.10.1.193:
    Packets: Sent = 4, Received = 0, Lost = 4 (100% loss),

Solution:

C:\>ping MASTER

Pinging master [10.10.1.1] with 32 bytes of data:

Reply from 10.10.1.1: bytes=32 time<1ms TTL=128
Reply from 10.10.1.1: bytes=32 time<1ms TTL=128
Reply from 10.10.1.1: bytes=32 time<1ms TTL=128
Reply from 10.10.1.1: bytes=32 time<1ms TTL=128

Ping statistics for 10.10.1.1:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
share|improve this question

1 Answer 1

When a remote computer connects to the Routing and Remote Access server by using a dial-up or a VPN connection, the server creates a Point-to-Point Protocol (PPP) adapter to communicate with the remote computer. The server may then register the IP address of this PPP adapter in the DNS or the WINS database.

When the Routing and Remote Access server registers the IP address of its PPP adapter in DNS or WINS, you may receive errors on the local computers when you try to connect to the server. You receive these errors because the DNS or WINS servers may return the IP address of the PPP adapter to computers that query DNS or WINS for the server's IP address. The computers then try to connect to the IP address of the PPP adapter. Because the local computers cannot reach the PPP adapter, the connections fail.

This knowledge base article applies up to Windows 2003 - so it may also apply to 2008, but it definitely describes your problem and offers a solution.

It's not recommended to run DNS and RRAS on the same server.

share|improve this answer
    
Thank you! I'm moving RRAS to secondary server and will test it. The KB article was also very helpful. –  Cyber Jul 2 '09 at 21:48
    
Oops. My fault. The secondary server is ALSO a DNS server. I will try to decipher all that Kb article, now ;) –  Cyber Jul 2 '09 at 21:51
    
The KB article will make it do what you want. This was a fairly common configuration back in my day-to-day SBS-installing days workin' for "the man". smile –  Evan Anderson Jul 2 '09 at 21:56

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.