Hi,
A lot of users in a remote site use port replicators (HP 3005pr), I've been informed by my colleagues that when a user in a remote site goes from port replicator A to replicator B, their IP address is updated with the IP associated with the port replicator device and DNS is not updated properly. The consequence of this is that the DNS record for the laptop client shows an invalid IP.
I've done some testing and I've observed that when a user plugs into a port replicator device, the MAC address of the port replicator is listed against the client lease (as shown by windows 2008 R2 DHCP console), rather than the MAC address of the windows 7 client. Furthermore, regardless of the laptop plugged into the port replicator, the laptop will always get the same IP if there's an active lease with the port replicator's MAC.
Each client is set to register their connection address in DNS. DHCP is running on Windows Server 2008 R2. DHCP scopes are configured to "Always dynamically update DNS A and PTR records", as well as "Discard A and PTR records when lease is deleted". Each subnet has 2 DHCP servers with over lapping scopes and exclusion ranges. Looking at the DHCP servers, I can't see any errors or warnings apart from the odd full DHCP scope, however in every case the second DHCP server has free addresses in the same subnet.
From my testing I've observed the following:
- When a laptop moves from one port replicator to another, the DNS entry for the laptop is updated initially on the PDC emulator, RID manager and infrastructure master first.
- Despite the DHCP server being a DC, it can take up to 3 minutes for it to update its own DNS record
- All DCs return the correct IP after 4 minutes
I'm wondering whether anyone else has experience of using these port replicators and has noticed similar behaviour?
Thanks