Details
-
Improvement
-
Resolution: Unresolved
-
Critical
-
5.1.1, 5.5.2
Description
This MB has been created to request a feature which helps isolate and manage problem indexes without affecting other indexes that are also being served from the index node(s).
When a cluster is serving multiple applications and these applications use indexes then when there is an issue which causes the indexer to restart then all indexes and services are affected detrimentally even if the issue is limited to one or more specific indexes/apps. What is being requested is that the indexer should be aware and be able to make a decision as to whether a full or partial restart of the service is required and if possible it should some how isolate problem indexes and it's causing affect. This should allow for other unaffected indexes and services to continue to work without interruption.