-
Type:
Story
-
Status: Closed (View Workflow)
-
Priority:
Major
-
Resolution: Fixed
-
Affects Version/s: None
-
Fix Version/s: 1.13.1
-
Component/s: None
-
Labels:
-
Story Points:3
-
Epic Link:
-
Sprint:N Sprint #6 - Platform & ISSU
The source of cluster instability is often livelocks in distributed primitives. These result from some inconsistency between the client and server session state. But clients should be able to detect this instability based on their failure to make progress and, rather than not recovering from the instability, recreating their session as a solution to session instability. This solution has the potential to suppress bugs, but it should go a long way towards ensuring the longevity of clusters.