Download PDF
Download page Enable and Configure Anomaly Detection.
Enable and Configure Anomaly Detection
Related topics:
Anomaly Detection must be enabled, but requires no configuration except to limit anomaly alerting. Enabling Anomaly Detection also enables Automated Root Cause Analysis.
Anomaly Detection and Automated Root Cause Analysis are available to only SaaS customers.
Enable Anomaly Detection
You need to enable Anomaly Detection separately for each application.
- In Alert & Respond > Anomaly Detection, select the desired application from the dropdown, and toggle Anomaly Detection ON.
After you enable Anomaly Detection, it takes 48 hours for Anomaly Detection and Automated Root Cause Analysis to become available. During that time, the machine learning models train on the business transactions in your application. Select Alert & Respond > Anomaly Detection > Model Training to view Business Transaction training status.
This table explains the training statuses of a Business Transaction.Status Meaning In Training Model training is in progress for the Business Transaction. Ready Model training is complete and the Business Transaction is healthy. Warning Model training is complete, but the Business Transaction has experienced one or more Warning level anomalies during the training period. Critical Model training is complete, but the Business Transaction has experienced one or more Critical level anomalies during the training period. Not Available Model training is incomplete and the Business Transaction is not visible to Anomaly Detection. The models continue training as long as Anomaly Detection is enabled. If traffic to a Business Transaction is interrupted for long enough duration preventing training that day, Anomaly Detection continues to function using the models from the previous seven days.
No machine learning models are trained for Business Transactions that have very low calls per minute (CPM), because the sample size will be so small that the resulting model will be unreliable.
Monitor Anomalies
- From Applications > Business Transactions, select any Business Transaction of interest.
- Click the Warning or Critical icon in the Health column.
A list of health rule violations and anomalies for the Business Transaction displays:
- You can view the lists of anomalies in multiple ways. Monitoring anomalies can reflect how you work with Splunk AppDynamics. Choose any of these options to open a detailed view that includes the results of Automated Root Cause Analysis.
- If you set up and validate tools for a tools team:
View the anomaly details from Alert & Respond > Anomaly Detection > Anomalies - If you monitor applications for an application operations team:
- From Applications > Events, filter Event Types to include anomalies
- From Applications > Events, filter Event Types to include anomalies
- If you set up and validate tools for a tools team:
- From Applications > Troubleshoot > Violations & Anomalies, filter Event Types to include anomalies
- From Applications > Troubleshoot > Violations & Anomalies, filter Event Types to include anomalies
You can configure policies to be triggered by anomalies, similar to the way you configure policies to be triggered by Health Rules.
Configure Anomaly Detection
By default, Anomaly Detection alerts you about the anomalies found in all the Business Transactions in your application. However, you can configure Anomaly Detection to surface only those anomalies within the combination of Business Transactions, severity level, and detection sensitivity that you specify. Do this if you prefer to see fewer and more narrowly focused alerts:
- Click Configure Anomaly Detection to open the configuration dialog.
- Select one of the following Business Transactions on which you want Anomaly Detection to alert:
- All Business Transactions in the Application (this is the default selection)
- Business Transactions within the specified Tiers
- These specified Business Transactions
- Business Transactions matching the following criteria:
- Starts With
- Ends With
- Contains
- Equals
- Matches Regular Expression
- Is in List
Is Not Empty
You can also select the NOT operator to reverse the criteria.
- Select one of the following severity levels:
- All Severities (includes both Warning and Critical)
- Critical
- Warning
In Detection Sensitivity, select one of the following levels:
Sensitivity Level Description High Use this level for business-critical services to ensure that no issue gets undetected in your environment. It triggers more alerts but with lower statistical confidence. Medium Use this level for services that are important to your business but not critical. By default, this sensitivity level is selected. Low Use this level for services that have low business impact and to avoid too many alerts. If you want to test anomaly detection in a non-production environment, select Yes, turn on test mode.
The test mode allows you to assess anomaly detection capabilities in non-production environments. In this mode, the anomaly detection accurately detects any performance issues even if metric data collection is low. You can use the test mode in your development or staging environments.
- Click Save to complete the configuration.