Good answer, I only came to add that OSQuery should be included in your tool box. Much easier to query for indicators of a problem, especially at scale, than it

Good answer, I only came to add that OSQuery should be included in your tool box. Much easier to query for indicators of a problem, especially at scale, than it is to grep through logs and reconstruct an event. This is still a detective tool that would not likely catch an incident in real time, so the preventative measures listed are still best practice, but it should be included as a tool to fall back on.