I installed the PAN user agent on a customers core DC and we are now having some issues. The traffic between the DC with the user agent and the firewall is pretty minimal and everything is working correctly. However we are seeing a lot of additional traffic ( consistent 1Mbps+) between their core DC (where the agent is installed), and their edge DCs that wasn't present before the install.
Has anyone else experienced this before? I have installed the agent on several customer networks without any issues, but they were quite a bit smaller domains.
Yes, I have seen that type of behavior a few times. The reason being that the agent is continuously reading the DC/Exchange logs that you point out in the agent configuration. The more complex the environment (number of users, number of polled DC's per agent) will have an impact on how much traffic that is generated. Unfortunately reading the security log is something that can consume quite a lot of bandwidth in some environments.
If possible I suggest you try adding one more agent and place it closer to the remote DC/Exchange server.
Its pretty odd considering I haven't seen this issue with any other customer's network that has had agents deployed. This customer however does have exchange where the other customers did not. Does exchange itself have something to do with it? The traffic between the agent and the firewall is really minimal but the traffic between sites is causing some issues. At this point the only solution I can think of is to install an agent on every DC. Not sure if there are any caveats with doing this...?
the User agent will need to read log files from each of the DC's since security logs are not sync'd between DCs. Over a wan link or a link which will have less bandwidth (remote site) i have seen where this communication can saturate the link. work around would to install agents in geo location.
The User Agent software can be quite chatty. If the User Agent is configured to talk to the other DCs, then it will query all of their user data. Based on my observations, the user agent software can generate between 1 to 2Mbps of traffic between itself and the DCs.
If any of those DCs are connected to the core via a T1 line, then it could easily swamp the WAN network.
The recommendation mentioned previously is the best. I would suggest installing the user agent software at the remote locations and then configure the PA firewalls to use the additional user agents.
Click Accept as Solution to acknowledge that the answer to your question has been provided.
The button appears next to the replies on topics you’ve started. The member who gave the solution and all future visitors to this topic will appreciate it!
These simple actions take just seconds of your time, but go a long way in showing appreciation for community members and the Live Community as a whole!
The Live Community thanks you for your participation!