hello,
if i have 2 nodes standalone elasticsearch working with scenario:
node-1 : have elasticsearch and logstash and logstash send logs let's called it "index-1"
node-2 : have elasticsearch and logstash and logstash sends logs let's called it "index-2"
now i am required to cluster these two nodes with a third new node, so i need to know they will replicate old data to each other or not ? and there will be data loss if i clustered them ?
thanks in advance
You can not join nodes that belong to different clusters, so what you need to do is:
- Stop Logstash so no indexing takes place
- Copy all data from node-2 to node-1. You can do this using snapshot and restore or by reindexing from remote.
- Update the config of node-1 so it can be part of a cluster. This may mean changing network interface etc.
- Stop node-2
- Delete all data from the data path of node-2
- Update configuration of node-2 so it is configured to join the node-1 cluster as a new empty node
- Restart node-2 and make sure it forms a cluster with node-1.
- Update index settings and index templates in case you want to have a replica shard for increased resiliency.
- You can now restart Logstash.
- Add node-3 as a new empty node to the cluster
hello, thanks for your help but I need to know how to do this part in the configuration below.
Have a look at the following links: