|
|
Jason Keltz wrote:
...
There are two VPN servers for load balancing/redundancy issues.
Each client on startup chooses a VPN server to connect to.
However, a client will only connect to one or the other.
In the event of a failure, a client will connect to the other VPN server.
Indeed if you are using two distinct VPN networks the load balancing
issue can be solved easily. The failover issue is the same with the
exception of a seamless hand over.
In terms of adding routes back to the VPN subnets, that is what I want
to do, but that is where there is a tricky problem...
When an individual client connects to either VPN server, it will get the
exact same IP. This is done so that software (like NFS) will be able to
basically handle a client moving from one VPN to the other if a VPN
server goes down...
Mhhh... NFS is stateless AFAIK. I don't know what happens when a NFS
mount goes down on the server and/or client, but I assume it will have
to be recovered anyway and cannot be seamlessly handed over. Most other
protocols will fail anyway.
cheers
Erich
____________________________________________
Openvpn-users mailing list
Openvpn-users@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/openvpn-users
|