What is pre-tunnel and post-tunnel latency?

[Geroge]

Hello!

I was wondering if anyone could shed some light on what exactly is meant by pre-tunnel and post-tunnel latency, which can be found under the GlobalProtect monitor.

Is it latency before the tunnel was created, and then after the tunnel was created?

Or does pre-tunnel latency + post-tunnel latency = tunnel latency?

What can it tell me about a clients connection? Low latency is obviously good, but when I see a pre-tunnel latency of 1200, can I point blame on the clients local internet connection, or elsewhere?

I can't find anything clear from Palo Alto on these terms, so any ideas on the subject would be very much appreciated!

[laurence64 · Accepted solution]

I would suggest that a pre-tunnel latency of 1200 is really quite bad and could lead to end user experience issues, the tunnel latency in my understanding is the latency to the endpoint when it logs onto the portal (pre-tunnel) and once the tunnel comes up (post-tunnel) the paths that the packets take across the internet will have a big bearing on the latency stats.

I would say the best use for this information would be with distributed gateways, for instance if you have a portal in the UK and a gateway in the US and the UK, a user in the US should use the gateway that is closest (least latency) and then use a faster connection through MPLS or SD-WAN back to corporate resources.