Problem Statement/Rationale

I found that after restoring data through mongodump and mongorestore, there is a probability of losing the {_id: "hashed"} index

Steps to Reproduce

1. Create a replica set (mongodb6.0)
2. Create an index on {_id : "hashed" }
3. insert some data
4. run mongodump like:  ./mongodump  --host="xxx" --username="xxx" --password="ixxx" --authenticationDatabase="admin" --authenticationMechanism="SCRAM-SHA-1" --db="xxx" --collection="xxx"
5. run mongorestore to restore  data to other replicaset : ./mongorestore  --host="xxx" --username="mongouser" --password="xxx" --authenticationDatabase="admin" --authenticationMechanism="SCRAM-SHA-1" 

Expected Results

Actual Results

The index on {_id: "hashed"} sometimes may be missing.

Additional Notes

I found after jira: https://jira.mongodb.org/browse/TOOLS-3108.

Previously, the _id index was filtered by index name.

But now the following method is used.

So the {_id: "hashed} index may be filtered out incorrectly, depending on its order in the indexes array.