Difference between Ingest Node and Coordinating Node?

Do Ingest node apply only to indexing from ingest pipelines?

And Coordinating Node to all other kinds of indexing?

Also client and Coordinating nodes are same right?

Assuming you are setting each node to be only the role you mention.





So all nodes are coordinating node correct?

Suppose I have 3 nodes. One ingest only, one data only, one master only.

Now if I send request to Ingest only node, will it act as coordinating node?

If I send to master node, will it act as coordinating node?

I assume the answers are yes to above questions. That would mean that all nodes are coordinating.Even if we turn them to specific type of node, they would still act as coordinating node if request is sent to them.

By default, all nodes play every role.

That's not really an ideal cluster. You have no HA, no redundancy.

If it's a request to an ingest pipeline, yes it will "coordinate" where it needs to be sent. But it won't accept a normal search or index request because it won't be attached to an ingest pipeline.

If it's master only, no.

Yeah I was just using that cluster as example to understand.

So what would happen to request which come to master only node or to Ingest only node?

I don't use ingest pipeline and logstash is pointed to Ingest only nodes and data is getting indexed.

It depends what the request is.

Let's take bulk index request or normal index request as example.

Check out

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.