I'm currently having all sorts of problems with ESX 4.1 and WIndows 2008 R2 x64. What I have is two ESX 4.1 hosts on a test network. When I install 2008 R2 into one host and promote it to a DC with DNS, and DHCP; I'm getting 404, 407, and 408 DNS errors which I never got with other versions of windows servers. I have tried different VM NICs; E1000, Vmxnet2, and Vmxnet3 to no avail. I then decided to create another DC in the other host and I am able to PING the first DC but when I try to DCPROMO it; it can not find the current DC. This is driving me crazy, any help or comments would be most appreciated. Thx!
I ran into a similar issue to this while working with Exchange 2010 on 2K8 R2 VM's. I had one VM that could not 'talk' to any other VM's but I could manage it and RDP to it. After trying all kinds of network troubleshooting from the Windows and networking perspectives I decided to just vmotion the VM to another host. This cleared up the issue. When I vmotion'ed it back to the original host the issue did NOT return.
This it the only time I have had this issue and the vmotion activity seemed to clear it up. I don't know if it was a bad ARP cache on the vSwitch or something odd with the network but it is gone now and I can't recreate it.
If none of the other suggestions help, I would try just deploying a new VM and promoting it to a domain controller. Domain controllers are expendable for the most part so when something goes wrong it's usually just as easy to just deploy a new one than try to fix an existing one, depending on the circumstances.
It does sound like it is something in the guest. When you say you disabled the firewall, you didn't simply stop the service did you? You should always disable the firewall from within the "Wijndows Firewall with Advanced Security" tool in Windows rather than just stopping the service. If you just stop the service it ends up blocking network connectivity to the server.
Start with the two VM on the same host (be sure that hosts have the same time, otherwise you get problem).
Join the second DC, before make dcpromo, this make sure that networking, firewall, clock, and so are fine.
Then make dcpromo.
After AD is replicated add the DNS Server role on the second DC.
I have the same issues with DNS (Same DNS errors) and no Internet access using a static IP. When I switch to DHCP (Using a scope on another server) no problem. I have a Citrix server and this secondary DNS and DHCP server on ESXi 4.1. All the others are physical servers in my network with a total of 10.
The DC promo followed SOP.
If I disable and enable the NIC I get a IP conflict notification but nothing else is using this IP, checked Spice Works and ran NetworkView to make sure I was not crazy. The MAC reported as duplicate IP does not exists. Yes, very strange.
I started troubleshooting based on the DNS issue and have done the following:
Reinstalled NIC, removed IP from DNS Interfaces, checked for other services using port 53 (None), MS KBs 956188, 812873, 929851. And some other posts related to the individual DNS errors. Disabled all firewall policies….private, public and domain (By GPO-Defaulted).
I even created a reservation for the IP I wanted on the other DHCP server and applied by changing the VM one to DHCP. This process does not give the IP conflict notification but does not allow internet access.
I have never used vmotion or can afford it. Any other ideas? I would like to resolve this on the current VM and not have to start over.