Configuring the removal of unavailable nodes from monitoring

Use this section to manage configuration variables for Elasticsearch monitoring policies.

To add a configuration variable

  1. On the Monitor Policies page, click Configuration and select Monitor Policies.
  2. Click the action menu of the policy that you want to modify and select Edit .
  3. In the Configuration Variables tab, click Add Configuration Variable.
  4. On the Add Configuration Variable page, in the Variable field, enter the variable path and name.
    For example, /ESK_TOP/hostList/<label>/removeNodeAfterNDiscCycle = <number of retries>
  5. Select the required Operation.
    • REPLACE: Create a new variable by using the REPLACE operation.
    • DELVAR: Deletes the variable and disables the task for which you created it. The value is deleted after you save the policy.
    • DELETE: Sets the variable value to an empty string. The value is deleted after you save the policy.
  6. In the Value field, enter the value of the variable.
    If you do not enter a value, the default value of the variable is used.

Monitoring can be configured for specific or global environments. To configure the removal of unavailable nodes from monitoring, you must use the following variables based on your requirements: 

Variable Description
/ESK_TOP/hostList/<label>/removeNodeAfterNDiscCycle = <number of retries>

Enables you to configure the removal of unavailable nodes in specific environments. Set thenumber of retries variable value in the pconfig file:

  • <label> is the name specified in the Label field during policy configuration while entering the Elasticsearch host details.
  • <number of retries> indicates the number of retries after which a node is removed from monitoring. The default value is -1, indicating that the unavailable nodes will not be removed from monitoring. You can set this variable to one of the following values:
    • 0: A node is removed from monitoring as soon it is determined to be unavailable.
    • Positive integer value:  The knowledge module attempts to connect to disconnected nodes for a specified number of discovery cycles. The knowledge module removes nodes from monitoring if it fails to connect to the nodes.
    • Negative integer value: The node is continuously monitored without being removed, regardless of the node's status.
/ESK_TOP/removeNodeAfterNDiscCycle =number of retries

Enables you to configure the removal of unavailable nodes from all environments.  Set the number of retries variable value in the pconfig file. The <number of retries> indicates the number of retries after which a node is removed from monitoring. The default value is -1, which indicates that the unavailable nodes are not removed from monitoring. You can set the variable to one of the following values: 

  • 0: A node is removed from monitoring as soon it is determined to be unavailable.
  • Positive integer value:  The knowledge module attempts to connect to disconnected nodes for a specified number of discovery cycles. The knowledge module removes nodes from monitoring if it fails to connect to the nodes.
  • Negative integer:  The node is continuously monitored without being removed, regardless of the node's status.
REMOVE_NODE_AFTER_N_DISC_CYCLE =number of retries

Enables you to configure the removal of unavailable nodes from all environments. Set the number of retries variable value as follows. 

The number of retries variable indicates the number of retries after which a node is removed from monitoring. The default value is -1, which indicates that the unavailable nodes are not removed from monitoring. You can set this variable to one of the following values:

  • 0: A node is removed from monitoring as soon as it is identified to be unavailable.
  • Positive integer value:  The knowledge module attempts to connect to disconnected nodes for a specified number of discovery cycles. The knowledge module removes nodes from monitoring if it fails to connect to the nodes.
  • Negative integer: The node is continuously monitored without being removed, regardless of the node's status.

Important: To configure this environment variable, you must stop and start the PATROL agent at the same terminal where the variable is configured.

 

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*