Re: [Openvpn-users] Curious port-forward artifact

2016-08-08 Thread RB
All-in-one to save chatter. Feel free to split if you need to respond. On Fri, Aug 5, 2016 at 2:34 PM, Gert Doering wrote: > Possibly rp_filter - "if looking up source address of an incoming packet > in the routing table leads to a *different* interface for egress, do not > permit this packet on

Re: [Openvpn-users] Curious port-forward artifact

2016-08-05 Thread debbie10t
https://forums.openvpn.net/viewtopic.php?f=6&t=22217#p63587 On 05/08/16 19:39, RB wrote: > I have a properly working server and client (both 2.3.11) passing > traffic mostly as expected, but have encountered some odd behavior I'm > hoping someone can enlighten me about. I've tested the below on

Re: [Openvpn-users] Curious port-forward artifact

2016-08-05 Thread Selva Nair
Hi On Fri, Aug 5, 2016 at 2:39 PM, RB wrote: > If, however, I advertise a default route over the tunnel, this works. > I also suspect that if I advertised a route to 2.2.2.2, it would also > work. > > Gert beat me to it, but the following may still help: Likely the packets are dropped by the ke

Re: [Openvpn-users] Curious port-forward artifact

2016-08-05 Thread Gert Doering
Hi, On Fri, Aug 05, 2016 at 12:39:01PM -0600, RB wrote: > Any guesses as to why this is? I've started to trace through the code > in an attempt to figure this out, but any pointers or answers would be > extremely welcome. Possibly rp_filter - "if looking up source address of an incoming packet i

[Openvpn-users] Curious port-forward artifact

2016-08-05 Thread RB
I have a properly working server and client (both 2.3.11) passing traffic mostly as expected, but have encountered some odd behavior I'm hoping someone can enlighten me about. I've tested the below on all three major platforms (as clients) with identical results. Say we have a client server at 1.