-
Type: Bug
-
Resolution: Fixed
-
Priority: Major - P3
-
Affects Version/s: None
-
Component/s: Replication
-
None
-
Replication
-
Fully Compatible
-
ALL
-
0
We re-connect all nodes at the end of the test so we can do a db hash check, but we don't wait for any rollbacks to finish, so it is possible we end up calling awaitReplication while a node is in rollback. This leads to us ignoring the node and not waiting for it to catch up. We should await secondary nodes after re-connection and before starting db hash checks.