Hi,
We have a problem in our organization. We are trying to add second node, to existing cluster, on the same machine. I know it's not a good solution, but we only do it for development purposes. The version that we are using is: 7.8
Our problem is that we are actually able to run two nodes on the same machine, but they don't detect each other - that is, when creating a node, each is created with totally new cluster ID, and they are not interconnected with each other.
When it comes to the configuration file, we only use the basic settings, as two nodes are running on the same machine, and there is no need to set up advanced stuff.
The parameters that we changed, is the node name, network host, we set up paths, node.max_local_storage_nodes: 2. Cluster name is the same for both nodes, and http ports are established automatically.
Is there any step that we skip that causes the two nodes do not find each other? Do you have any ideas? Thanks in advance!
This parameter is deprecated, don't use it, it's trappy because it requires the nodes to race against each other when starting up. Instead give each node a different value for path.data.
You have not set cluster.initial_master_nodes correctly, it should be a list of the node names:
As it says, it needs to discover nodes node-1 and node-2 but it's only discovered node-1. You've edited this message to remove details like the discovery config which is often the source of this problem -- you're making it really quite difficult to offer useful help!
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.