How does the end user know to connect to that specific IP you're seeing in SmartLog?
Most likely it's a DNS name they are typing into a web browser or an app is using.
If you know that DNS name (I.e. The end user can tell you), then you can follow the SK's I've linked to.
If you don't know this DNS name, you create a host object with the IP address you're seeing in SmartLog and put it in a bypass rule.
Obviously if the IP changes (or there are multiple IPS) then you'll have to create another object and add it to the rule.
If you can do a packet capture on the client DNS traffic, you may be able to figure out what DNS name the client is trying to connect to and use the SK'S I've linked.
Hope that's clear.