Pix 7.2(1) RDP sessions dropping with a TCP Reset-O

[ccioperations]

I've got some users connecting over RDP to internal windows hosts, they're getting regular intermittent timeouts, i've checked my logs and it looks like the connections being torn down with a tcp reset-O on the end.

Jul 24 13:15:56 10.0.1.1 Jul 24 2006 12:18:16: %PIX-6-302014: Teardown TCP connection 605272 for outside:External-Office/1283 to Dev:internal-host/3389 duration 0:08:45 bytes 152790 TCP Reset-O

Jul 24 13:15:56 10.0.1.1 Jul 24 2006 12:18:16: %PIX-6-106015: Deny TCP (no connection) from External-Office/1283 to ext-internal-Host/3389 flags PSH ACK on interface outside

That's as much as i can get from the logs, unfortunately not enough to resolve the problem, can anyone shed some light on the tcp reset-o and suggest what's happening, and even better, a fix! :)

Thanks in advance

Richard

[valconix]

Are you saying RDP does actually work ok, but getting the odd timeouts or slow connections within the rdp session? Reset-O is saying, the outside end disconnected the session, possibly due to slow response to the server? or loss session connection. What is the speed of the connection?

[ccioperations]

Yeah initially it's fine, but after a variable amount of time it'll disconnect, from the user's end it'll flash the network icon on the rdp session as if the connection was lost.

These problems only started when I upgraded my pix's to 7.2(1) from 7.0. I'm wondering if since the upgrade they're less tolerant to a troubled connection.

My servers are in London on a 100meg line and the users are in central Russia.... so yeah, it's entirely possible it's down to the distance between. But I can't change that variable. So i'm hoping I can tweak something at my end to fix it.