When you set up the Jive platform for on-premise
HA search, you will perform several steps in a specific sequence.
Note: As of Jive 7.0, the search index rebuild process has been improved so that you no
longer have to rebuild the search index on one node and then copy it to all of the other
search nodes. In versions 7.0+, the ingress replicators automatically send rebuild
traffic to all search nodes. Because of this change, all of the search nodes must be
available before starting a search rebuild. This ensures that the search index on the
search service nodes are always consistent.
In the following topics we will walk you through an example of an HA search configuration
setup that uses the following ports and hosts. Your configuration may vary depending on
2 search nodes: search01.yourdomain.com, search02.yourdomain.com, port 30000
2 ingress replicator nodes: ir01.yourdomain.com, ir02.yourdomain.com, port
1 haproxy node: haproxy.yourdomain.com, load balancing the search nodes on port
20000 and load balancing the ingress replicator nodes on port 19000
Setting Up the HA Search Proxy
To configure a redundant HA search environment, you will need a proxy server to load balance requests to each search server and ingress replicator.
Services Directory for HA Search
The search service relies on a file named serviceDirectory.json. This file should be identical for all of the search servers (not the ingress replicators) in your HA configuration.