Does anyone here have any experience running the Crowdstrike Falcon Sensor in their Splunk environment? I've found the following: https://docs.splunk.com/Documentation/Splunk/8.2.5/ReleaseNotes/RunningSplunkalongsideWindowsantivir... but it references on-access AV, and Crowdstrike is a behavioral AV and that likely isn't totally applicable. I have a case open with Splunk with this same question but I wondered if the community had any experience; do's/don'ts; best practices; etc. My gut is that I won't see a substantive performance impact but I'd love to have a little more knowledge before I start deploying the agent.
Trying to search for this online has proven neigh impossible since CS-->Splunk integration is very common and almost all the search hits focus on ingesting CS logs, not actually running the agent on a Splunk environment.
For reference I have a modestly sized distributed architecture with three search-heads and three indexers (not clustered) in addition to a deployment and multiple forwarders.
Did you ever got any answer on this?
I never did, no, but I went forward with configuring this myself. I run a test environment for Splunk, so I was able to confirm that there didn't seem to be any adverse affects from running the sensor on my hosts. Since deploying it in production I've had zero issues and zero detections of any kind. I created a dedicated host group with tags to manage my Splunk environment separately (if necessary) but I haven't had to. I have pretty default linux sensor settings and it has been working fine. I'm happy to share any specific configurations I have in place if you need any guidance.