Hi there,
I'd like to talk about an old problem again - the "unidentified network"-problem when using OpenVPN-Tap-Adapters.
Currently I am trying to use sophos connect again and run into the same issue with the network not identifying after connection. That results in a non working DNS for the internal clients. Connections using IP are working. I guess the problems cause is, that we use the same DNS zone internally as externally.
I know about lots of possible solutions by searching for "unidentified network and openvpn or tap adapter like using a bogus route, changing registry keys, using secpol etc but none of them are comfortable, some don't even work for me (e.g. I can't find that registry setting for ndiswan as is does not seem to get used by the new sophos connect client). My windows firewall is disabled.
Currently I have found another "solution" that I do not understand. I just set the setting for adapter metric not to automatic but to 1. Then the adapter gets quickly identified in my test machine, although the internal routes for my connected network still have a higher metric value than the local standard gateway. I also do not know if this will consistently work for other clients and again this just seems to be some kind of workaround than a real explanation.
So can somebody tell me, what really causes this problem and how to get easily over it?
And if not, why does this metric setting solve the problem? Easiest solution would be currently using powershell:
Get-NetAdapter -InterfaceDescription "Sophos TAP Adapter" | Set-NetIPInterface -InterfaceMetric 1
Nevertheless - this sucks!
Cheers
Marcel
This thread was automatically locked due to age.
