Config Churn is for SCOM similar to us processing old news – like the news of some days ago - and never getting into a state where we get to the point to process the latest news from BBC/CNN/Fox News, just because we can’t keep up with all the changes happening around us. Like an overload and never ever reaching a status quo.
For SCOM this situation is just as bad since it never gets into a state where it knows the current status of all monitored servers. The monitored servers themselves are way behind in processing their configuration as well, which may result in running old configurations (Rules/Monitors/Discoveries/Workflows and the lot). All this can lead to the SCOM Management Servers becoming stale, which is noticed by these servers turning gray in the Console. Ouch!
How to fight it?
Since every Management Group differs in size, dimensions (the amount of secondary Management Servers, Gateway servers), imported and tuned MPs (are there any home brewed among them?) it’s hard to tell exactly when Config Churn will take place. Changes are however, at a certain moment in time there will be Config Churn to some degree.
This is also the reason why there aren’t any Monitors/Rules in place which will tell you when Config Churn is taking place. Simply because it can manifest itself in many different ways. Yes, there are certain EventIDs in the OpsMgr event log which point into that direction. True. But per Management Group it really differs.
Therefore it’s important to keep track of the OpsMgr event log of the RMS on a regular basis in order to see all is well. This doesn’t have to take much of your time, ten minutes at the most. Just run the event log once a week and be done with it.
Is there a KB article for it?
But how to recognize Config Churn? And how to fight it? And how to identify the culprits? There are multiple good resources to be found on the internet but on different places. Gladly KB2603913 solves this issue and has grouped them all together. It tells you what Config Churn is, how to recognize it, who the usual suspects are, how to identify them, and what to do about it.
So go check it out this KB article since it’s really a good one with lots of good and solid information.