Documentation Plugins AppDynamics data source Query editor

AppDynamics query editor

The AppDynamics query editor allows you to build queries for AppDynamics data in dashboards, Explore, and alerting. The query editor supports the following query types:

• Metrics - Query application performance metrics from the AppDynamics Metrics API.
• Analytics - Query analytics data using ADQL (AppDynamics Query Language) from the Analytics Events API.
• Health - Query health rule violations for an application.
• Events - Query application events such as deployments, errors, and policy violations.
• Tiers - Query tier information for an application.
• Metric Names - Browse the metric hierarchy for an application.

Select a query type using the radio buttons at the top of the query editor. For general information on Grafana query editors, refer to Query editors.

Key concepts

If you’re new to AppDynamics, here are key terms used in this documentation:

Metrics query type

To create a Metrics query, select Metrics from the query type radio buttons. The Metrics query editor queries application performance metrics from the AppDynamics Controller REST API.

Raw edit mode

By default, the metric field uses segment-based selection where you navigate through the metric hierarchy one level at a time. Click the edit icon (pencil) next to the Metric field to switch to raw text input mode, which allows you to type or paste a full metric path directly. Click the checkmark icon to apply changes when using raw edit mode.

Configure labels

Customize how series are labeled in your visualizations using the Labels drop-down below the main query fields.

• Full Path - Uses the full metric path as the label. For example: Overall Application Performance|Average Response Time (ms).

• Segments - Builds the label from specific segments of the metric path. Segment indexing starts at 1. For example, with a metric path of Errors|mywebsite|Error|Errors per Minute and segments 2, 4, the label shows mywebsite|Errors per Minute.

• Custom - Combines free text with aliasing patterns to include metric metadata:

  • {{app}} - The application name.
  • {{n}} - The nth segment from the metric path.

  For example, with the metric path Overall Application Performance|Average Response Time (ms) and the custom label {{app}} MetricPart2: {{2}}, the result is myApp MetricPart2: Average Response Time (ms).

Analytics query type

To create an Analytics query, select Analytics from the query type radio buttons. Analytics queries use ADQL to query AppDynamics analytics data.

The query editor provides a code editor with autocomplete suggestions for fields, tables, and template variables as you type your ADQL query.

Analytics query examples

The following examples demonstrate common ADQL query patterns.

Count transactions grouped by name, filtered by a template variable:

SQL Copy

SELECT distinct(transactionName), count(*) FROM transactions WHERE transactionName IN (${transactionName:doublequote})

Aggregate response times into a time series by user experience:

SQL Copy

SELECT series(eventTimestamp, '1m'), userExperience, count(*) FROM transactions

Query the slowest transactions over the selected time range:

SQL Copy

SELECT transactionName, max(responseTime) AS maxResponseTime FROM transactions ORDER BY maxResponseTime DESC LIMIT 10

Query application logs for errors:

SQL Copy

SELECT eventTimestamp, MSGTY, MSGTXT, Application FROM application_logs WHERE MSGTY = 'E'

Query browser performance records:

SQL Copy

SELECT pageUrl, avg(domReadyTime) AS avgDomReady, count(*) AS total FROM browser_records ORDER BY avgDomReady DESC

Filter to the dashboard or alert time range:

SQL Copy

SELECT series(eventTimestamp, '1m'), count(*) FROM transactions WHERE eventTimestamp >= $__from AND eventTimestamp <= $__to

Set the Limit field to control how many records are returned (default 100). The query editor suggests a maximum of 1000; the backend doesn’t enforce that cap. For aggregation queries like count(*) or avg(), the limit applies to the number of result rows, not the number of raw records scanned.

Time macros

The plugin interpolates the following time macros in Analytics queries to epoch milliseconds from the dashboard or alert time range. Use them in ADQL filters so queries work in dashboards and in alerting:

Example:

SQL Copy

SELECT count(*) FROM transactions WHERE eventTimestamp >= $__from AND eventTimestamp <= $__to

To use these macros in alert rules, refer to AppDynamics alerting.

Health, Events, Tiers, and Metric Names query types

The Health, Events, Tiers, and Metric Names query types use form-based editors to query the AppDynamics Controller REST API. Each type provides specific input fields for its API endpoint.

The Health, Events, and Tiers query types include a Use dashboard’s time picker toggle. When enabled (the default for new queries), the query uses the dashboard time range and hides start-time and end-time. When disabled, set a fixed range with time-range-type, start-time, end-time, and duration-in-mins as needed. Metric Names queries don’t include this toggle.

In alerting, the toggle is hidden and the alert evaluation window is used as the time range.

Time-range-type values are BETWEEN_TIMES (default), BEFORE_NOW, BEFORE_TIME, and AFTER_TIME:

• BETWEEN_TIMES uses the dashboard or alert time range.
• BEFORE_NOW requires duration-in-mins.
• BEFORE_TIME requires duration-in-mins and end-time.
• AFTER_TIME requires duration-in-mins and start-time.

Times are in milliseconds since epoch.

Health

Health queries retrieve health rule violations for an application. Use a table visualization.

To list open and recent health rule violations:

1. Select Health from the query type radio buttons.
2. Select your application from application_id.
3. Leave Use dashboard’s time picker enabled.
4. Use a Table visualization.

Events

Events queries retrieve application events such as deployments, errors, and policy violations. Use a table panel, or overlay the same query as annotations. For annotation setup and a list of common event types, refer to AppDynamics annotations.

eventtype and event-types are different fields. Use event-types to choose which events AppDynamics returns. Leave eventtype at the editor default unless you have a reason to change it. The editor marks summary and comment required.

To list deployment events in a table:

1. Select Events from the query type radio buttons.
2. Select your application from application_id, or type ${application}.
3. Fill summary and comment.
4. Leave eventtype at the default.
5. Select APPLICATION_DEPLOYMENT from event-types.
6. Select INFO from severities.
7. Leave Use dashboard’s time picker enabled.

The application_id drop-down lists applications, not dashboard variables. Type ${application} as a custom value rather than selecting a numeric application ID.

Tiers

Tiers queries list tiers in an application. When you set a time range, SaaS Controllers can limit results to tiers that were alive in that window. On-premises Controllers return all tiers.

To list tiers:

1. Select Tiers from the query type radio buttons.
2. Select your application from application_id.
3. Leave Use dashboard’s time picker enabled.
4. Use a Table visualization.

Metric Names

Metric Names queries browse the metric hierarchy. They don’t return time-series values. Use this type to discover valid metric paths for Metrics queries or template variables.

To list metrics under overall application performance:

1. Select Metric Names from the query type radio buttons.
2. Select your application from application_id.
3. Set metric-path to Overall Application Performance.
4. Use a Table visualization.

Use cases

The following examples demonstrate common use cases for the AppDynamics data source.

Monitor application response time

Query the average response time for an application to track performance trends:

1. Select Metrics from the query type radio buttons.
2. Select your application from the Application drop-down.
3. Navigate to Overall Application Performance|Average Response Time (ms) in the Metric selector.
4. Set Labels to Custom and enter {{app}} - Response Time.

Display a single metric value

Use Roll-up and a Stat panel to display a rolled-up value for the time range:

1. Select Metrics from the query type radio buttons.
2. Select your application from the Application drop-down.
3. Navigate to Overall Application Performance|Calls per Minute in the Metric selector.
4. Set Aggregation to Sum.
5. Toggle Roll-up on.
6. Use a Stat or Gauge visualization.

Monitor a business transaction

Query calls per minute for a business transaction. Use * in a path segment to match all transactions:

1. Select Metrics from the query type radio buttons.
2. Select your application from the Application drop-down.
3. In raw edit mode, enter Business Transaction Performance|Business Transactions|*|Calls per Minute.
4. Set Labels to Segments and enter 3 so the series name is the business transaction.

Track error rates

Query error metrics to monitor application health:

1. Select Metrics from the query type radio buttons.
2. Select your application from the Application drop-down.
3. Navigate to Overall Application Performance|Errors per Minute in the Metric selector.
4. Set Aggregation to Max to see peak error rates.

Compare metrics across applications

Use a template variable for the application to compare the same metric across multiple applications:

1. Create a template variable with the query Applications.
2. Select Metrics from the query type radio buttons.
3. Select ${application} from the Application drop-down.
4. Navigate to the desired metric path.

When multiple applications are selected, the plugin generates a separate query for each application and displays them as individual series.

List health rule violations

Use the Health query type to show violations in a table:

1. Select Health from the query type radio buttons.
2. Select your application from application_id.
3. Leave Use dashboard’s time picker enabled.
4. Use a Table visualization.

Analyze business transactions

Use an Analytics query to analyze business transaction volume and performance:

SQL Copy

SELECT transactionName, count(*) AS total, avg(responseTime) AS avgResponseTime FROM transactions ORDER BY total DESC

Identify slow transactions

Find transactions that exceed a response time threshold:

SQL Copy

SELECT transactionName, responseTime, eventTimestamp FROM transactions WHERE responseTime > 5000 ORDER BY responseTime DESC

Set Limit to 50 to retrieve the top 50 slowest transactions.

Monitor user experience distribution

Visualize how user experience is distributed across transactions:

SQL Copy

SELECT series(eventTimestamp, '5m'), userExperience, count(*) FROM transactions

This query returns a time series grouped by user experience category (NORMAL, SLOW, VERY_SLOW, STALL, ERROR), suitable for a stacked bar or time series visualization.