#Resolved After a lot of thinking and poking around I found the issue. Under System -> Routing (System:Gateway): I found that the default gateway was set incorrectly. Many thanks for your time. Mike

#Resolved After a lot of thinking and poking around I found the issue. Under System -> Routing (System:Gateway): I found that the default gateway was set incorrectly. Many thanks for your time. Mike

Thanks. It was a transient problem, I guess. It's working now.

Not entirely sure what you mean by 'doom2 source port master'. The last time I played Doom 2 it was over a very long serial cable!  ;) What requirements does that server have? Does it require you forward ports? UPNP? Do you have a link? Steve

I don't think it's latency.  Logged on from the LAN last night… still no refreshing. :(

Can you post your RTT (pings should be fine) between those offices? I have noticed that IPSec is especially sensitive to congestion when there are long round trip times and bandwidth is higher (you can read a little about it if you Google "Bandwidth Delay Product"). There is a whitepaper that sheds a bit of light on the IPSec part here: http://www.academia.edu/694268/_TCP_in_the_IPSEC_environment To work around this, you may seriously consider implementing an OpenVPN solution to replace IPSec. pfSense has an excellent implementation of OpenVPN point to point and can handle failover (via CARP).

Still no info on the "hang" problem, but I'm betting on the issue where you need to disable SMP in the BIOS for install and then reenable it later.  That seems to hit a decent number of Dell servers.

No problem mate :) Glad to help because this community helped me a lot :)

@denask: Any security gains are from having additional devices on a separate subnet anyway, I guess. If you have two interfaces bridged into one subnet you can still apply firewall rules at those interfaces to filter traffic to some extent. So there some security gains if you need them. I would also just attach it to the switch though unless you really need filtering. Steve

Thank you !!

@marcelloc: sip phones are on lan too? If so try to configure a heartbit on your sip servers. Depends on what kind of SIP servers he has on the network. Not all have heartbeat feature. I recommend using DNS server for load balancing. Use round robin feature.. you can add both SIP servers in DNS and the request goes in a round robin fashion. You can have multiple SIP servers in load balancing this way.

Errors are gone now, transfer speeds up to: 70 Mbytes/s. PCIe ASPM was to blame, hope it helps some one :) Thanks stephenw10, I really appreciate it!

I suspect this is fixed by this commit from jimp a couple of hours ago: https://github.com/pfsense/pfsense/commit/c1e48e9b40cb08884a6b4848ef95ec27b6311549

thanks, now it works

It's not quite the same as accessing a modem where the PPPoE interface doesn't use the actual ethernet interface. In this case the interface connecting to the ubnt device is already in use so a virtual IP must be used. Steve

@NOYB: I think the difference between the two ways that would be most likely to impact would be the NAT/Firewall settings I use to restrict WoL begin accepted from only a few static source IP addresses.  Guessing your mobile device wouldn't have a static IP address.  So the NAT/Firewall rule would have to accept any, or at least a much larger range, of IP addresses.  But I think your NAT/Firewall rule already allows any address to accommodate the mobile device.  Both ways should work with mobile device so long as the NAT/Firewall is configured to accommodate it. It makes sense. I will start to monitor my cell phone IP just to see if its going to change. Thank you!

Thanks for giving me some interesting things to think about and to look at. I will tinker around a bit, never considered looking at power settings because all cores showed 50c. Which now that i think about it is prob on the low side for 100%.

No problem - Just a lucky guess.