Monitoring Cluster Nodes that Host Partitions

You can find out which nodes in a Data Fabric cluster are being used for topics in a stream by running the command maprcli stream topic info. The nodes are listed in the servers field.

The guts utility can show you whether there are any I/O bottlenecks on these nodes. This utility can also show you whether there is any capacity on other nodes in the cluster that you can take advantage of by creating additional partitions for topics.

To run this utility, issue this command after logging into the Data Fabric cluster that you want statistics for:

/opt/mapr/bin/guts

You can also use the guts utility to show only these statistics from HPE Data Fabric Streams:

Table 1.
Name Description
mpr The number of RPCs from HPE Data Fabric Streams producers to the server.
mpm The number of messages that have been published to the server.
mpMB The total size in MB of the messages that have been published to the server.
mlr The number of RPCs from HPE Data Fabric Streams consumers to the server.
mlm The number of messages that have been read from the server.
mcl The number of concurrent RPCs from consumers to the server.
mlMB The total size in MB of the messages that have been read from the server.

To see these statistics, run this command: 

/opt/mapr/bin/guts streams:all
NOTE
These statistics are for the most recent sample period at the time the command is run, and are not cumulative. Sample periods are one second.