This section describes the configuration entities created by the dynatrace-service on the Dynatrace tenant when it receives a sh.keptn.event.monitoring.configure event. This makes it easy to configure your Dynatrace tenant to fully interact with the Keptn installation.
To trigger automatic configuration, execute the following CLI command where <PROJECT_NAME> is the name of the associated Keptn project:
keptn configure monitoring dynatrace --project=<PROJECT_NAME>
To enable or disable the creation of the following entity types, please see Configuring automatic generation of Dynatrace entities.
Once processing of the configure monitoring event is complete, the dynatrace-service sends a sh.keptn.event.configure-monitoring.finished event with a summary of the operations performed.
When dynatraceService.config.generateTaggingRules is set to true, the dynatrace-service will create tagging rules for keptn_service, keptn_stage, keptn_project, keptn_deployment tags. For example the rule for keptn_project is created as follows:
{
"name": "keptn_project",
"rules": [
{
"type": "SERVICE",
"enabled": true,
"valueFormat": "{ProcessGroup:Environment:keptn_project}",
"propagationTypes": [
"SERVICE_TO_PROCESS_GROUP_LIKE"
],
"conditions": [
{
"key": {
"attribute": "PROCESS_GROUP_CUSTOM_METADATA",
"dynamicKey": {
"source": "ENVIRONMENT",
"key": "keptn_project"
},
"type": "PROCESS_CUSTOM_METADATA_KEY"
},
"comparisonInfo": {
"type": "STRING",
"operator": "EXISTS",
"value": null,
"negate": false,
"caseSensitive": null
}
}
]
}
]
}When dynatraceService.config.generateProblemNotifications is set to true, the dynatrace-service will try to create a problem alerting profile named Keptn with rules for AVAILABILITY, ERROR, PERFORMANCE, RESOURCE_CONTENTION, CUSTOM_ALERT and MONITORING_UNAVAILABLE that trigger problem notifications after 0 minutes for all entities in all management zones. If an alerting profile is already available it is not overwritten.
The alerting profile is then used to create a webhook named Keptn Problem Notification to send problem events to Keptn using the event API. The webhook has the following form:
{
"type": "WEBHOOK",
"name": "Keptn Problem Notification",
"alertingProfile": "<ALERTING_PROFILE_ID>",
"active": true,
"url": "<KEPTN_ENDPOINT>/api/v1/event",
"acceptAnyCertificate": true,
"headers": [
{
"name": "x-token",
"value": "<KEPTN_API_TOKEN>"
},
{
"name": "Content-Type",
"value": "application/cloudevents+json"
}
],
"payload": "<PAYLOAD>"
}Values are set for <ALERTING_PROFILE_ID>, <KEPTN_ENDPOINT> and <KEPTN_API_TOKEN>. The actual template, added as <PAYLOAD>, has the form:
{
"specversion": "1.0",
"type": "sh.keptn.events.problem",
"shkeptncontext": "{PID}",
"source": "dynatrace",
"id": "{PID}",
"time": "",
"contenttype": "application/json",
"data": {
"State": "{State}",
"ProblemID": "{ProblemID}",
"PID": "{PID}",
"ProblemTitle": "{ProblemTitle}",
"ProblemURL": "{ProblemURL}",
"ProblemDetails": {ProblemDetailsJSON},
"Tags": "{Tags}",
"ImpactedEntities": {ImpactedEntities},
"ImpactedEntity": "{ImpactedEntity}",
"KeptnProject": "<PROJECT_NAME>"
}
}The value of <PROJECT_NAME> is set to the Keptn project being configured.
If a problem notification named Keptn Problem Notification already exists it is overwritten.
When dynatraceService.config.generateManagementZones is set to true, the dynatrace-service tries to create a management zone for the project and for each stage it contains. The project management zone, named Keptn: <PROJECT_NAME>, contains services tagged with keptn_project: <PROJECT_NAME>, whereas each stage management zone, named Keptn: <PROJECT_NAME> <STAGE_NAME>, contains services tagged with keptn_project: <PROJECT_NAME> and keptn_stage: <STAGE_NAME>. If a management zone with the same name already exists, it is not overwritten.
When dynatraceService.config.generateDashboards is set to true, the dynatrace-service creates (or overwrites) a dashboard called <project-name>@keptn: Digital Delivery & Operations Dashboard. The dashboard contains some basic infrastructure monitoring tiles for the health of hosts, CPU load and network status, as well as a default quality-gate comprised of service health, throughput, failure rate and response time.
When dynatraceService.config.generateMetricEvents is set to true, the dynatrace-service tries to create custom alerts for each service on each stage in the project based on the associated SLIs and SLOs.