[SLING-4590] topology console reporting a inconsistent state of topology for isolated cluster node

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: Discovery Impl 1.1.0
• Fix Version/s: Discovery Impl 1.1.8
• Component/s: Console
• Assignee: Stefan Egli
• Reporter: Michael Tarantino
• Created: 08/Apr/15 16:13
• Updated: 30/Sep/15 05:44
• Resolved: 22/Sep/15 14:35

Description

After a split because of a heart beat timeout, the topology console is reporting an inconsistence state of the topology. The screenshot below shows 4 instances while the local cluster node is isolated from the rest of the cluster.

Sling id	ClusterView id	Local instance	Leader instance	In local cluster	Announced by instance
0d08c47d-6a1b-48d2-87b2-f88813da03f5	f07aec64-4289-4b07-becc-dfdd76178f04	false	false	remote	(changing)
9d2bad89-bba7-47d8-bcf7-090d7989a747	f07aec64-4289-4b07-becc-dfdd76178f04	false	false	remote	(changing)
c1d45df4-8703-4ada-85db-ae597cda8032	f07aec64-4289-4b07-becc-dfdd76178f04	true	true	local	n/a
c26b9027-c367-425e-8fa4-8b595f476473	f07aec64-4289-4b07-becc-dfdd76178f04	false	false	remote	(changing)

The issue might be located between 'topology.getLocalInstance().getClusterView()' and 'clusterViewService.contains' which seem to differ with the isolated cluster mode.