Validators are configured on sources. You can add more than one validator on a source to monitor calculated metrics and statistics for different use cases.

To configure a new validator,

1. Navigate to the source where you want to add the new validator.
2. Click + New validator to start the configuration wizard.
3. Choose the Validator Type you want to configure on this source. For more information, see Validator Types.
4. The Metric steps in the configuration depend on the selected validator type, and includes specifying fields and filters for validation, specifying windows to aggregate the data, segmenting the data for more granular analysis.
5. (Optional) Check Initialize using historical data to prime the anomaly detection algorithms.
6. Define the Threshold to use for detecting anomalies and triggering incident notifications. For more information, see About Thresholds.
7. Complete the Validator Details,
   1. Enter a Name and (optional) Description to help explain what the validator is used for.
   2. (Optional) Add Tags, which you can use to categorize the validator.
   3. Assign an Owner, who can be responsible for handling incident notifications from this validator.
8. Click Continue to create the validator.
   After the validator is created, click View validator to open its details page. It will take a few minutes for data to start populating the graph and table on this page--depending on the polling schedule and if you manually backfill the validator.

Initialize using Historical Data

All validator types except for the metadata validators support backfilling, or initializing the validator with historical data. Starting the validator with historical data primes the anomaly detection algorithms, so that you will start to see incidents detected on the past data. As you triage these incidents, you retrain the algorithms for future incident detection.

Backfill occurs when you start the source for the first time, if historical data exists. If you want to load historical data to new validators on a source that is already running, you can select the Initialize using Historical Data option.

📘

Note

The amount of historical data, or Lookback period, is defined on the Source-level, but the validator will not use historical data unless the Initialize using Historical Data option is selected. For more information, see Configuring a Source.

Validator Ownership

The owner you assign to the validator will be automatically assigned to incidents detected by that validator and for routing incident notifications. You can update and assign a new owner to the incident group during investigation and triage. For more information, see Managing Incident Groups.