Based on the logs collected, mongodb query planner seems to be doing replanning several times. 

The logs show that the query where there is either a match or a sort operation on the _id field causing mongodb query planner to do replan. Here is the sample log:

planSummary: IXSCAN { _id: "hashed" } cursorid:23334124130458 keysExamined:205 docsExamined:101 fromMultiPlanner:1 replanned:1 replanReason:"cached plan was less efficient than expected: expected trial execution to take 4 works but it took at least 40 works" numYields:9

This is sometime causing heavy CPU utilisation on MongoDB server with many such queries getting into the replanned stag. Can someone from MongoDB confirm if this is a known behaviour where query or sort based on the _id? This is also seen when the mongodb has multiple indexes matching to the query and _id is one of the either for the sort or match part.