theOpenVPNbibo
Hi there, this is my first post...
I just ported an long time running "Remote Access Server" which uses OpenVPN for remote access from VMWare Server 1.x to Hyper-V (Windows 2008 R2, the OS in the VM is Windows XP).
It runs in general but since the migration from VMWare to Hyper-V I have many, real many connection aborts after some minutes with normal operation.
I tried several things (like disabling time synchronisation of the integration services and so on) but nothing helped until now.
In the OpenVPN log I found the following messages when the connection breaks down:
MULTI: multi_create_instance called
Re-using SSL/TLS context
...
TLS Error: TLS key negotiation failed to occur within 60 seconds (check your network connectivity)
TLS Error: TLS handshake failed
...
After reading all the FAQs and hints concerning this fault messages (firewall, router, ...) I just changed the normal network card of the Hyper-V and now using the legacy network adapter.
Until now the behavior is total different and this seems to be the solution for that problem!
I do not know what the difference between these two types of adapters is but the normal adapter of the integration services seems to generate the described problems...
Any comments or similar experiences?
Mimiko
As there is not much info about lan configurations, config files, logs, I suppose that you have a mtu problem. Try lowering MTU packet size in OpenVPN config.
Native network driver is working fine under Hyper-V with windows XP VM. Legacy network card are slow.
theOpenVPNbibo
Thank you for your quick answer!
I simply thought it would be helpful to reduce the given information to bring it to the point:
Simply changing the network adapter solved the problem...
Of course I will give the whole information if needed, but... how can changing the network adapter solve a MTU based problem?
