I have configured an OpenVPN bridge on a linux box that connects small subsidiaries running OpenWRT in a Linksys router.
Central:
Code: Select all
192.168.1.x <-----> Switch <------> [LX OpenVPN eth1, tap0] | [eth0, 3.x] <-----> [router] <----> (internet)
( ......... internet ................)
<-----> Linksys OpenWRT (OpenVPN bridged) <-----> 192.168.1.x
<-----> Linksys OpenWRT (OpenVPN bridged) <-----> 192.168.1.x
<-----> Linksys OpenWRT (OpenVPN bridged) <-----> 192.168.1.x
<-----> Linksys OpenWRT (OpenVPN bridged) <-----> 192.168.1.x
<-----> Linksys OpenWRT (OpenVPN bridged) <-----> 192.168.1.x
Everthing works perfect but the access from the headquater PCs to the internet. Yes, we can access the internet, but the bandwidth is less than the half available bandwidth. Avaiable: 35 MBit, Actual bandwidth from inside the headquater: 15 MBits. If I run several bandwidth tests from the OpenVPN router, everthing is ok.
So there must be some trouble with the routing or bridging. It seems, that the traffic will be slowdowned by the bridge.
Is it possible, that this bridging configuration slows down the access to the public internet in that dimension? I hoped, that the internal 192.168.1.x traffic will be fowwarded from eth0 to eth1 bypassing the bridge.
I hope my description was clear enough. Perhaps everything is ok and i have to live with this behavior. But in the past our headoffice had an bridged linksys too and it runs (almost) perfect. Not until a new subsidiary sets boundaries.
Thank you