OpenVPN on pfSense will send more than enough keep alives to keep that up, and OSPF's hello packets would be more than enough as well. For what you show to happen it really would have to lose connectivity for 60+ continuous seconds between the sites.

Amazing.

leighno5 - Did you get this working? I am having the same issue. If you did, can you post what you did to make it work? I came up with a boatload of IP addresses for Hulu, but with them using Akamai, it's hard to pinpoint the correct ones.

I did notice if I check the No Pull box, start Hulu, then uncheck and apply, Hulu continues to work.

ops dont know why i said both are pfsense routers as one is and ones a draytek so thats why i created vm openvpn servers one at both sites behind the routers

I'm not sure that this is a pfSense problem, but probably more of a general VPN VoLTE problem. I have my VPN configured to not route all traffic through the VPN, it only sends the traffic on my work network through the VPN. Public internet still goes out on the regular public gateway.

I've got AT&T and called them today to try to troubleshoot the issue. The lady was surprising knowledgeable and actually reached out to their Tier 3 when she was on the phone with me. He did a call trace and said it looked like a device issue not a network issue so they referred me to Samsung. They offered to conference the call with them, but I had to jump on a conference call.

If you have any success or further input on what you found out about this I'd love to hear it.

My guess is that the call signaling still goes out over the regular voice network and they just reroute the audio payload over the data network. My phone will initiate the call, but it usually drops before the other side ever even rings; at most I get 1 unanswerable ring.

Actually it was a misconfiguration on one of the clients, it was missing the remote subnet option. I did not need to add the client-to-client option on the server side, it looks like when selecting peer to peer it's already there by default.

Everything seems to be working now, thanks

192.168.101.101 is the server address in your case. You cannot push it to a client.

This works perfectly as what I want.

Thank you viragomann

The wizard is pretty much IDIOT proof, yet seems like every other day we have someone trying to use a user cert for the server…

wizardservercert.png
wizardservercert.png_thumb

I'm really late here but i can tell you expressvpn would be great tool for you solution :) its mos tsecure vpn service provider list on mostsecurevpn.com

Hi, ref 1), advice on how would I do the correct routes? Thanks.

Tunnelblick works perfect with with config  exported via
Inline Configurations: Others

Well i solved the problem.
Rebuild the VPN using http://blog.stefcho.eu/building-site-to-site-connection-with-openvpn-on-pfsense-2-0-rc1-with-pki/

that Rebooted Both Firewall….

Than to redirect the internet throught the vpn just chect the related boxes and and create the nat routes...

Glad you go it sorted… and happy I could point out how simple it was ;)

Hi again, thanks for your comments. I have tried saving the server config again with no luck. I am not using subnet 192.168.80.X por any other purpose.

I think the problem might have anything to do with the fact that the client gets assigned ip address 255.255.255.255 instead of 192.168.80.X (see the image).

Any other suggestions?


Found the problem, I'm also using "User Auth", and I had "auth-nocache", apparently after one hour I have to renter the user and password! I removed "auth-nocache" and the problem is solved. Is this a security hole? Allowing the cache of the user credentials?

Thank you all
Best Regards