I am having an issue with machine name resolution. There is a LogMeIn Hamachi VPN involved but it may or may not have anything to do with the problem.
- Machine A, intranet, Windows Server 2008 R2 64-bit, running as the VPN gateway
- Machine B, intranet, Windows 7 Pro SP1 64-bit
Machine C, intranet, Windows XP Pro SP3 64-bit
Machine D, home PC connecting using VPN, Windows 7 Pro SP1 32-bit
Machine E, home PC connecting using VPN, Windows 7 Home Premium SP1 64-bit
We do not use any domain (very small network)
I have done the following tests (when I say VPN I mean machine connecting in using VPN to one of the internal machines)
All intranet pings by name or IP work correctly (between machines A, B, C).
All VPN pings by IP work correctly.
All VPN pings by name from D to A/B fail intermittently. Sometimes the name resolves correctly then a minute later it will fail then it will work again.
All VPN pings by name from E to A/B fail always.
All VPN pings by name from D/E to C work always.
In both the above cases the failures are because the name resolves to the intranet IP rather than the VPN virtual IP. For example the intranet IPs are 192.168.1.x whereas the VPN assigns 10.40.1.x
For example from machine E ping A will resolve to 192.168.1.37 (intranet IP) when it should resolve to 10.40.1.37 (VPN alias IP)
Accessing network shares using windows explorer by name always works from outside the VPN. It seems to be straight-forward TCP/ICMP services such as intranet web pages, ping, subversion servers etc that do not work.
It would also appear that it might be something to do with the OS of the target machine as the newer OSs such as Windows 7, Windows Server 2008 R2 do not work but Windows XP does.
Can anyone suggest either a possible cause/solution or any steps I should be taking to narrow down the issue? For example it would useful to try and trace exactly where it is getting the incorrect IPs from in the failing case versus the correct IP in the successful cases (DNS/NetBios/Something else).
