This check collects metrics for your Cassandra cluster that are not available through jmx integration. It uses the
nodetool utility to collect them.
This check is packaged with the Agent, so simply install the agent.
Create a file
cassandra_nodetool.yaml in the Agent’s
init_config: # command or path to nodetool (e.g. /usr/bin/nodetool or docker exec container nodetool) # can be overwritten on an instance # nodetool: /usr/bin/nodetool instances: # the list of keyspaces to monitor - keyspaces:  # host that nodetool will connect to. # host: localhost # the port JMX is listening to for connections. # port: 7199 # a set of credentials to connect to the host. These are the credentials for the JMX server. # For the check to work, this user must have a read/write access so that nodetool can execute the `status` command # username: # password: # a list of additionnal tags to be sent with the metrics # tags: 
When you run
stackstate-agent info you should see something like the following:
Checks ====== cassandra_nodetool ----------- - instance #0 [OK] - Collected 39 metrics, 0 events & 7 service checks
cassandra_nodetool check is compatible with all major platforms
|Percentage of data available per keyspace times replication factor|
shown as percent
|Replication factor per keyspace|
|Node status: up (1) or down (0)|
|Percentage of the data owned by the node per datacenter times the replication factor|
shown as percent
|Amount of file system data under the cassandra data directory without snapshot content|
shown as byte
The Cassandra_nodetool check does not include any event at this time.
cassandra.nodetool.node_up: The agent sends this service check for each node of the monitored cluster. Returns CRITICAL if the node is down, otherwise OK.