-
Type: Task
-
Resolution: Unresolved
-
Priority: Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
Labels:
-
Service Arch
-
0
We have sys-perf workloads that spin up a lot of threads to do a heavy workload against a mongodb cluster. This occasionally hits a connection storm and crashes nodes in the server. We should start implementing some policies in our sys-perf project that makes use of our workload management mechanisms (like PM-3613) to see if we can stabilize the cluster during these test runs.