DSE Multi-Instance commands

Commands to configure and use multiple DataStax Enterprise nodes on a single host machine.

Commands to configure and use multiple DataStax Enterprise nodes on a single host machine.
Note: DSE Multi-Instance commands are supported only on package installations.

dse add-node

Synopsis
$ sudo dse add-node [command_arguments]
To show the command line help for any command:
$ sudo dse add-node --help
The following table describes the command arguments for the dse add-node command. When optional parameters are absent, the default value remains unchanged. With DSE Multi-Instance, the dse add-node command arguments define some properties that are otherwise set in the cassandra.yaml file. When an optional parameter is omitted, the default value is not changed.
Command arguments Description
nodeId Adds the specified node to an existing DataStax Enterprise cluster, where the specified nodeID is the node name.

The specified node name is automatically prepended with dse- so that the resulting node ID is dse-nodeId. For example, if you specify node1, the resulting node name is dse-node1.

[--cluster=clustername] The name of the DataStax Enterprise cluster that the node belongs to. Only non-whitespace values are supported.
[--jmxport=jmx_port] The JMX port for the node.
[--listen-address=IP_address] The IP address or hostname that Cassandra binds to when connecting to other nodes.
[--rpc-address=IP_address] The IP address or hostname that Cassandra binds to for RPC requests.
[--cpus=number] The number of Cassandra cores.
[--max-heap-size=size] Set the Java heap size. The size must be specified in megabytes. If you omit MB the size is interpreted as megabytes.These commands are equivalent:
$ dse add-node -n node1 --max-heap-size=800
$ dse add-node -n node1 --max-heap-size=800MB
[--data-directory=directory] The root directory for storing data on each node. Default: /var/lib/dse-nodeId/data
[--hints-directory=directory] The hints directory for each node. Default: /var/lib/dse-nodeId/hints
[--commit-directory=directory] The commit log directory for each node. Default: /var/lib/dse-nodeId/commitlog
[--cdc-directory=directory] The CDC raw data directory for the node. Default: /var/lib/NODEID/cdc_raw
[--saved-caches-directory=directory] The saved caches directory for each node. Default: /var/lib/dse-nodeId/saved_caches
[--logs-directory=directory] The root directory for all of the logs for each node. Default: /var/log/dse-nodeId
[--tomcat-logs=directory] The log directory for tomcat logs. Default: logs-directory/tomcat where logs-directory is defined with --logs-directory.
[--hadoop-logs=directory] The log directory for Hadoop logs. Default: logs-directory/hadoop
[--hive-logs=directory] The log directory for Hive logs. Default: logs-directory/hive
[--pig-logs=directory] The log directory for Pig logs. Default: logs-directory/pig
[--mahout-logs=directory] The log directory for Mahout logs. Default: logs-directory/mahout
[--spark-worker-directory=directory] The data directory for Spark Worker for the node. Default: /var/lib/dse-nodeId/spark/worker
[--spark-worker-cores=cores] The maximum number of cores used by Spark executors.
[--spark-worker-memory=memory] The maximum amount of memory used by Spark executors. Specify unit of measure with k (kilobytes), m (megabytes), g (gigabytes).
[--spark-shuffle-service-port port_num] The Spark shuffle service port. Default: 7737
[--spark-local-directory=directory] The local directory for Spark Worker for each node. Default: /var/lib/dse-nodeId/spark/rdd
[--spark-log-directory=directory] The log directory for Spark Worker for the node. Default: /var/log/dse-nodeId/spark/worker
[--unix-username=name] The UNIX user that owns the node configuration. Default: cassandra.
[--unix-group=group] The UNIX group that owns the node configuration. Default: cassandra.
[--tomcat-logs=directory] The directory for Tomcat server logs. Default: logs-directory/tomcat
[--hadoop-logs=directory] The log directory for Hadoop server logs. Default: logs-directory/hadoop
[--rack=rack_placement --dc=datacenter_placement]

rack_placement is the rack placement for the node.

datacenter_placement is the datacenter placement for the node.

[--seeds=IP_address1,IP_address2,...] A comma-separated list of IP addresses of the nodes to be used as seed nodes.
[--search] Enable DSE Search on the new node. Default: not enabled
[--analytics] Enable DSE Analytics (Spark) on the new node. Default: not enabled
--help Send command descriptions to standard output.

dse list-nodes

$ sudo dse list-nodes
This command takes no arguments and lists the nodes that are configured for the DSE Multi-Instance host machine.

Since the default DataStax Enterprise node is called dse, the dse list-nodes command always returns at least the dse node, even if nodes were not added with the dse add-node command.

The location of the dse.yaml file depends on the type of installation:
Installer-Services /etc/dse/dse.yaml
Package installations /etc/dse/dse.yaml
Installer-No Services install_location/resources/dse/conf/dse.yaml
Tarball installations install_location/resources/dse/conf/dse.yaml

dse remove-node

Synopsis
$ sudo dse remove-node nodeId [--yes]
To show the command line help for any command:
$ sudo dse remove-node --help
The following table describes the command arguments for the dse remove-node command.
Command arguments Description
nodeId Required. Removes the specified node to an existing DataStax Enterprise cluster, where the specified nodeID is the node name. Because the node name is always prepended with dse- the command works if you specify dse-nodeID or just nodeID.
[--yes] Confirm node deletion. Files are deleted and are not recoverable. If you do not specify --yes on the command line, you are prompted to confirm node deletion.
--help Send command descriptions to standard output.