I have half a dozen sg650s in regional areas where I have to rely on multiple adsl connections for some redundancy.
We have lots of users who connect to a hosted accounting system that works via Citrix metaframe (Now Citrix Xenapp) which is a remote desktop.
The snapgear now pings the gatweay that each connection sees to decide if the connection is up and this seems quite reliable. Originally it had been set up to ping the telstra dns server but that seems really flaky and causes the connection to fail over unnecesarily.
The issue is that when the connection fail sover and sends a user's data out on one of the other connections citrix loses the connection and you have to go back and login and pick up your session again. Browsing and downloading and odbc queries can all cope with the connection changing seamlessly to the user but citrix doesn't seem to be able to.
I have got the connections failing over well and I use policy routes to send some users over one connection or other but I am not an expert by any means..
I also understand that this is quite possibly more about citrix than the utm netgear side of things but I was hoping someone could point me in the right direction!
Thanks in advance.
Failover on the SG UTM device is NOT stateful....as such connection need to be reestablished.
What you are seeing is expected behavior
Thanks Ross- excellent.
Is there no way around this other than spending thousands on a stateful router? You guys are clever- just make some new firmware