-
Type: Task
-
Resolution: Fixed
-
Priority: Major - P3
-
Affects Version/s: None
-
Component/s: Sharding
-
Fully Compatible
-
Sharding 2021-04-19, Sharding 2021-05-03
-
2
Prior to SERVER-54513, it was possible for a donor/recipient machine to not exist by the time the coordinator decides to abort - and the coordinator needed participants to report that they had seen the operation was being aborted.
With SERVER-54513, the coordinator should know whether participant state machines have been made and whether it needs to wait for the participants to acknowledge the abort before completing.
- is related to
-
SERVER-54000 Make errors propagate from the ReshardingCoordinator to participants
- Closed
-
SERVER-54513 Create new resharding recipient state kAwaitingFetchTimestamp
- Closed
- related to
-
SERVER-53611 Handle errors on donor before donor document is initially inserted
- Closed
-
SERVER-56288 Don't wait for participant machine creation in resharding_abort_command.js
- Closed