-
Type:
Improvement
-
Resolution: Fixed
-
Priority:
Major - P3
-
Affects Version/s: None
-
Component/s: None
-
Catalog and Routing
-
Fully Compatible
-
CAR Team 2025-09-01
-
200
-
1
-
🟩 Routing and Topology
-
None
-
None
-
None
-
None
-
None
-
None
Similar to how gFindChunksOnConfigTimeoutMS overrides the defaultConfigCommandTimeoutMS (30s default) for the chunks namespace on config server queries, we should explore doing something similar for config.shards lookups.
config.shards lookups should be pretty fast given that the response is just the list of shards, and although unlikely, in the event the request fails due to network issues or config node going down while refreshing the ShardRegistry, this might cause an unnecessarily long stall before the operation is retried.
Stalls in the ShardRegistry can be pretty bad as they cause user operations to stall too.
- causes
-
SERVER-110098 Increase default timeout for findShardsOnConfigTimeoutMS for ubsan slow variants
-
- Closed
-
- related to
-
SERVER-110755 Make moveCollection thread in resharding_skip_cloning_and_applying.js retry on NetworkInterfaceExceededTimeLimit error
-
- Closed
-