Manual configuration

Performance monitoring tools can also be configured manually by the user.

Important: If you are using IBM Spectrum Scale™ 4.1.1 or later version, the performance monitoring tool gets automatically configured. This will automatically override any manual changes you try to make to the configuration. If you wish to change an automated configuration to a manual one, follow the steps given in the Removing an automated configuration section in Automated configuration.

When configuring the performance monitoring tool manually, the installation toolkit sets up a default set of sensors to monitor on each node. You can modify the sensors on each individual node.

The configuration file of the sensors, ZimonSensors.cfg, is located on each node in the /opt/IBM/zimon folder. The file lists all groups of sensors in it. The configuration file includes the parameter setting of the sensors, such as the reporting frequency, and controls the sensors that are active within the cluster. The file also contains the host name of the node where the collector is running that the sensor must be reporting to.

For example:

sensors =
{
        name = "CPU"
        period = 1
},
{       name = "Load"
        period = 1
},
{
        name = "Memory"
        period = 1
},
{
        name = "Network"
        period = 1
        filter = "eth*"
        # filters are currently ignored.
},
{
        name = "Netstat"
        period = 1
},

The period in the example specifies the interval size in number of seconds when a sensor group gathers data. 0 means that the sensor group is disabled and 1 runs the sensor group every second. You can specify a higher value to decrease the frequency at which the data is collected.

Whenever the configuration file is changed, you must stop and restart the pmsensor daemon by using the following commands:

Issue the systemctl stop pmsensors command to stop (deactivate) the sensor.
Issue the systemctl start pmsensors command to restart (activate) the sensor.

Some sensors such as the cluster export services sensors run on a specific set of nodes. Other sensors such as the GPFSDiskCap sensor must run on a single node in the cluster since the data reported is the same, independent of the node the sensor is running on. For these types of sensors, the restrict function is especially intended. For example, to restrict a NFSIO sensor to a node class and change the reporting period to once every 10 hours, you can specify NFSIO.period=36000 NFSIO.restrict=nodeclass1 as attribute value pairs in the update command.

Some sensors, such as VFS, are not enabled by default even though they have associated predefined queries with the mmperfmon query command. This is so because the collector might display performance issues of its own if it is required to collect more than 1000000 metrics per second. To enable VFS sensors, use the mmfsadm vfsstats enable command on the node. To enable a sensor, set the period value to an integer greater than 0 and restart the sensors on that node by using the systemctl restart pmsensors command.