Customizing a Monitoring Template

Customizing a Monitoring Template

There are a number of pre-existing monitoring templates, and if any of them satisfies your needs you can skip to the next section.

All templates are text files located under the directories

  • /etc/collectd.templates/snmp
  • /etc/collectd.templates/ipmi

according to the protocol used to collect the monitoring data, either SNMP or IPMI.

Any new file in these directories will automatically appear in the user interface, as shown in the next section, step 6.

SNMP Template

To create a new SNMP template, login as root, make a copy of one existing template, and then start editing it.

The SNMP template file has two types of subsections:

  • Data
    • One entry per datapoint, each identified by a unique ID.
  • Host
    • One single entry, defines the SNMP parameters, the collecting interval, and which data points are to be collected.

If you want to exclude data points you are not interested in, just remove it from the Collect option.

If you want to add additional datapoints, copy an existing Data section and adjust the options:

  1. Type
    The type of the measurement. Possible types are:
    • 1. temperature
    • 2. fanspeed
    • 3. humidity
    • 4. counter
    • 5. percent
    • 6. timeleft
    • 7. voltage
    • 8. current
    • 9. power
    • 10. apparent_power
    • 11. power_factor
    • 12. frequency
  2. Table
    true: if the OID is part of a table.
    false: if the OID corresponds to a single value.
    This needs to be defined before Instance or Values
  3. Instance
    If Table is true: A SNMP OID prefix that will be walked to retrieve a list of names that will be associated with the corresponding values. For example, in a PDU this could be the outlet name.
    If Table is false: The name [of the instance] that will be associated with the value, as a string.
  4. InstancePrefix
    Optional. A string to the prepended to the Instance, enclosed in double quotes.
  5. Values
    If Table is true: The SNMP OID prefix that will be walked to retrieve a list of values.
    If Table is false: The SNMP OID used to retrieve a single value.
  6. Scale
    Optional. A decimal value to be multiplied to the value retrieved before persisting it.

Except for Table and Scale, all others must be between double quotes.

IPMI Template

The ‘discover’ template for IPMI will automatically discover all the sensors available.

The template will have only one subsection, Host, and the options of interest are:

  1. AuthType
    The authentication type for the IPMI protocol. The default is to negotiate the strongest one. Possible values are:
    • 1. none
    • 2. md2
    • 3. md5
    • 4. straight
  2. Privilege
    The privilege level for the IPMI protocol. The default is admin. Possible values are:
    • 1. callback
    • 2. user
    • 3. operator
    • 4. admin
  3. Sensor
    Selects sensors to collect or to ignore, depending on IgnoreSelected. May be defined multiple times, each one selecting one sensor.
  4. IgnoreSelected
    If true, will not collect that for the sensors selected by Sensor.
    If false, will only collect the sensors selected by Sensor.
  5. Scale
    If any of the sensors report a measurement multiplied by a factor, you may want to add this. Its format is:
    Scale “<sensor name>” <multiplier>

Except for IgnoreSelected and the Scale’s multiplier, all others must be between double quotes.

    • Related Articles

    • NodeGrid Monitoring Aggregation Settings

      Here is how to configure a monitoring aggregation. 1) Make sure you have a monitoring license installed. If you don't have one, add one via System > License 2) Make sure you have the devices you want to aggregate configured and monitoring is enabled. ...
    • Enabling Monitoring

      Perform the steps below to enable monitoring for a device, using the interface of your choice, either the CLI or the web. Using the CLI Go to the device’s management settings: cd /settings/devices/<device_name>/management/ Configure monitoring for ...
    • Exploring Data Points

      This section will describe how you can verify that we are actually persisting the data collected and to learn more about the data being collected. This knowledge will be helpful in the Creating a Visualization article. First, make sure that you have ...
    • Considerations on Visualization

      Additional Considerations There is a lot of flexibility on creating visualizations, and the described steps cover only a small subset of what is possible. Other type of visualization is the area chart, which is useful for stacking measurements for ...
    • Creating a Visualization

      Perform the steps below to create a visualization, graphing the data collected. 1. Click on Visualize: 2. Select a Line chart: 3. Select a search source by clicking on ‘From a new search’. 4. Select ‘logstash-*’ as index pattern. 5. Select the data ...