-
Type: Bug
-
Resolution: Done
-
Priority: Major - P3
-
Affects Version/s: None
-
Component/s: Networking, Sharding
-
None
-
ALL
This leads to the case where a server goes down and comes back up, and hours later a socket exception occurs b/c the connection was never used previously.
EDIT:
This is addressed by SERVER-9041 (proactively checking idle connections in the pool) and SERVER-11681 (no longer holding thread-local connections in mongos between requests)
- is duplicated by
-
SERVER-9041 proactively detect broken connections detected by the network
- Closed
- related to
-
SERVER-9041 proactively detect broken connections detected by the network
- Closed
-
SERVER-11681 Mongos cleanup and integration of write commands
- Closed