The _id field for time-series collections is actually a user specified field. Unlike for other collections, we can't perform a sharded time-series single delete just with _id field in the query. The error message says "A single delete on a sharded collection must contain an exact match on _id (and have the collection default collation)" even though the query is on the _id field. We should remove this part of the comment for time-series collections.
- related to
-
SERVER-80104 Sharded time-series buckets should allow deleteOne against `_id`
- Closed