is there any guidelines for how much disk space we should be provisioning for fleet for an onprem deployment? Does it stabilize at a certain point? or does it just grow infinitely, and we are expected to prune somehow at certain intervals?

Hey @mason kemmerer, Is difficult to give some numbers for this as most of our deployments are ec2/cloud-based and have not really explored "bare metal" setups at this time. Some load-testing info that might answer some questions can be found in our engineering handbook. Guessing you're talking about logging destinations? It is dependent on frequency and number of endpoints, along with the number of policies/queries/etc. We've had some clients configure dumps to

stdout

and then they leverage

syslog

for retention and filtering. Some other community members might be able to share their infrastructure and configurations to give an idea of the size you might need.

thanks I'll take a look at this