Understanding Network Discover Cluster Port usage
Article ID: 277401
Updated On:
Products
Issue/Introduction
When configuring Network Discover Cluster, there is a need to limit or restrict the recommended ranges of default ports
Resolution
The Discover Cluster uses these port ranges to ensure high availability of the socket binding process.
These ranges can be reduced or configured to use one port, but be aware that in the case of the worker node, the same client connection port must be used as the client connection port configured in the data node.
The Cluster Discovery Port Range is used by 1st/master data node to discover other data nodes in the same cluster. At present, we only support single data node, so at this time it's not being used, but mandatory to initialize the data node.
Having a configuration mapped as follows -> StartPort=EndPort=SamePort <- should function in Cluster Discovery Port Range configuration.
From data node scope, the master node listens on the port for all other data nodes to connect.
As for the Client Connection Port Range, it is the channel between data-node and worker-node, which they share any data required for exchange over the network. The more worker nodes that are used and number of scans run, the more advantageous to have default port ranges configured for availability. This is dependent on the scale of the configuration.
Feedback
