Details
-
Bug
-
Resolution: Works as Designed
-
Major - P3
-
None
-
4.0.5
-
None
-
ALL
-
Description
If mongos needs to hit multiple shards then the query stalls but if I run the query on both shards independently they finish instantly. I have a sharded collection on r,u. Additionally, if I run a command on the mongos that only needs to query a single shard it finishes instantly as well.
"conn34" is the affected connection in the logs. It looks like it returns from a single node very quickly and the other server is the slow one. I attached the operation from that server by running db.currentOp(true).
Attachments
Issue Links
- is related to
-
SERVER-39141 The range deleter should retry on write conflict exceptions
-
- Closed
-
- related to
-
SERVER-44092 Orphan documents impact IXSCAN performance
-
- Closed
-