-
Type:
Task
-
Resolution: Unresolved
-
Priority:
Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
None
-
Cluster Scalability
-
7
-
None
-
None
-
None
-
None
-
None
-
None
-
None
We have a few test suites/concurrency tests with stepdowns, but we don't have the equivalent for abort. Abort is tested in a few targeted tests and we got a couple of BFs from them. However, because of the frequency and limited configuration, it takes a long time for it to catch issues.
The idea is to add a concurrency test variant that aborts resharding and verifies that it can cleanup properly and not get stuck. We can possibly combine this with fcv change testing, since it implicitly aborts any resharding that is running.