Node Type Configuration
We have 4 Instances in Prod. We have configured like below.
3 instances - WebUser
1 instance - BackGroundProcessing.
How do I configure Search and Stream?
1. If I configure the Search on the 1 node, when the application restarts, if the re-indexing happens on one of the 3 nodes, Search wont work. Correct?
2. Can I configure all 4 nodes as Stream Node?
Ford Motor Company
US
You can label any given jvm/pod more than one node type.
I would suggest, you can limit the "Search" Node type limited to One jvm/pod
& Stream Node type would be either 3 or 5 Nodes/pods in a multi jvm/pod cluster.
BNYMellon Technology
IN
@SUMAN_GUMUDAVELLY thanks for your response. The inderxing happens randomly on available nodes, right. When if the Indexing happens on node that is not configured for Search Node Type?
Updated: 1 Apr 2021 1:15 EDT
Ford Motor Company
US
If we do not label any particular JVM with Search in a given Cluster, Pega will start the Search Service in the available JVM.
General Troubleshooting Guide for Search Issue: https://community.pega.com/knowledgebase/articles/pega-platform-troubleshooting/troubleshooting-pega-platform-search
In case if you label more than one JVM dedicated for Search then Pega will choose the First JVM which joined the Cluster with Node Type Search and it will start building the Indexes [Data, Work, Rule], depending on the configuration.
Also, please look at the troubleshooting videos posted by GCS which explains in detail about Search configuration and associate troubleshooting
https://community.pega.com/video-library/global-search-and-reindexing-issues-part-1 https://community.pega.com/video-library/global-search-and-reindexing-issues-part-2