In a customer environment I had the issue that the Domino Server crashed when a repeating Meeting Event was scheduled for a very long time ( in my case it was a weekly meeting for about 20 years ) and where a ressource was booked.

After opening a case the HCL support could reproduce this issue and sent the following description:

based on the call stack it looks like the R&R autoprocessing code was simply logging an action to the $CSTrack item. As RnRMgr creates each new accepted repeating meeting instance it logs the action to $CSTrack on the request for diagnostic purposes.

Making the $CSTrack item growing too large when they generate the repeating reservations in the database.

This issue is reported under SPR JPAIC5CHNJ and will hopefully be fixed in the upcoming release. If you have the same problem you can refer either to the SPR or to case CS0281893.