We have a Windows Domain Controller(DC) that creates lots of security events. We are monitoring wineventlog://security and have noticed that, on a regular basis, we get no events. We know the events are there because we can see events we missed in the evt files that are kept. The OS has a policy to rotate the log when it gets to a certain size, and we are thinking this may be when we stop getting events. This happens on all of our DCs, but not at the same time. Any experience with high volume DCs and logs rolling causing a problem?
@bstimely Assuming you have a high performing IDX at the recipient end, i would perform the following analysis to make the changes
Check the pipelines at the UF, HF (If there is an intermediate HF) and IDX and see that there is no blocks causing the delay
If the clogging is in the HF, try to play around maxthrougput and parallelingestionpipelines ; also ensure that the HF performance is very minimal and it is utilizing less than 50% of resources so that there are no overhead when you enable these configurations
Play around the maxthroughput of UF and see if it allows forwarding of events from the UF faster.