At the moment we don't have a DNS server configured on our network, as we have always used, and still do, local hosts files to indicate the host name and IP address of servers. This, as I'm sure you will agree, not an ideal situation, especially when it comes to resolving host names across the network.
A DNS problem (or what I think is one!) has been shown up by remote users connecting to our network via VPN. The Watchguard firewall/vpn issues the user with an IP address in its reserved range, therefore allowing it to connect to the network. The user than uses RDC to connect to a Terminal Services server. Upon connecting to the session, a drive mapping is created so that the user can see their local drive shares in the remote session, therefore being able to transfer files between the server and their laptop.
The problem I have is that the host name of the laptop cannot be recognised by the server until the IP address of the laptop is pinged, therefore placing it in the ARP table. Once in the table, you can then use the UNC path names to access drive shares without any problems.
My question is, what are the implications to the network structure upon installing the DNS service on a Windows 2000 server, that is not the Primary Domain Controller. (The PDC is a WinNT server only just capable of running Exchange) Would this solve the problems with device names not being recognised?
Also, would it be better to move to Active Directory, which incorporates DNS. I'm assuming this may be the better solution, now that the majority of our clients are Windows XP Pro.
I apologise if I don't make complete sense, as this is all relatively new to me. Any help is greatly appreciated.
Darren
A DNS problem (or what I think is one!) has been shown up by remote users connecting to our network via VPN. The Watchguard firewall/vpn issues the user with an IP address in its reserved range, therefore allowing it to connect to the network. The user than uses RDC to connect to a Terminal Services server. Upon connecting to the session, a drive mapping is created so that the user can see their local drive shares in the remote session, therefore being able to transfer files between the server and their laptop.
The problem I have is that the host name of the laptop cannot be recognised by the server until the IP address of the laptop is pinged, therefore placing it in the ARP table. Once in the table, you can then use the UNC path names to access drive shares without any problems.
My question is, what are the implications to the network structure upon installing the DNS service on a Windows 2000 server, that is not the Primary Domain Controller. (The PDC is a WinNT server only just capable of running Exchange) Would this solve the problems with device names not being recognised?
Also, would it be better to move to Active Directory, which incorporates DNS. I'm assuming this may be the better solution, now that the majority of our clients are Windows XP Pro.
I apologise if I don't make complete sense, as this is all relatively new to me. Any help is greatly appreciated.
Darren
