Open engineering-issue-sync-app[bot] opened 1 month ago
For more information and best practices, please visit confluence.
Provide monitoring endpoints to SaaS customers so they can monitor application-level metrics and analyze the performance of clusters.
Camunda SaaS platform does not expose monitoring endpoints
As a C8 SaaS user, I need to be able to monitor my cluster resource utilization using my own monitoring system.
As a C8 SaaS user, I can activate\/deactivate and secure the monitoring endpoint for my C8 Cluster.
As a C8 SaaS user, I can include IP addresses in the Allow list that will apply to Monitoring endpoints.
As a C8 SaaS user, I can get a sample Grafana Dashboard that can provide monitoring example for C8 Cluster
We will require that Monitoring tools can authenticate with bearer_token against the Management API Endpoint. (true for Prometheus and Datadog)
A new scope 'monitoring' will govern access when creating API keys.
API Endpoint where monitoring is exposed \/prometheus\/\<component>
Need to consider how to discover this endpoint from the Admin API
How the cost of Egress will affect our financial
Open Question:
Can we support OIDC authorization?
Do we have to filter out data from the monitoring endpoint?
Do we proxy existing endpoints or do we create a service monitor\/collector that will pass on metrics to the user endpoint?
:robot: This issue is automatically synced from: source
For more information and best practices, please visit confluence.
Value Proposition Statement 🚀
Provide monitoring endpoints to SaaS customers so they can monitor application-level metrics and analyze the performance of clusters.
Camunda SaaS platform does not expose monitoring endpoints
User Stories 🧑🚀
As a C8 SaaS user, I need to be able to monitor my cluster resource utilization using my own monitoring system.
As a C8 SaaS user, I can activate\/deactivate and secure the monitoring endpoint for my C8 Cluster.
As a C8 SaaS user, I can include IP addresses in the Allow list that will apply to Monitoring endpoints.
As a C8 SaaS user, I can get a sample Grafana Dashboard that can provide monitoring example for C8 Cluster
Implementation 👷
We will require that Monitoring tools can authenticate with bearer_token against the Management API Endpoint. (true for Prometheus and Datadog)
A new scope 'monitoring' will govern access when creating API keys.
API Endpoint where monitoring is exposed \/prometheus\/\<component>
Need to consider how to discover this endpoint from the Admin API
How the cost of Egress will affect our financial
Open Question:
Can we support OIDC authorization?
Do we have to filter out data from the monitoring endpoint?
Do we proxy existing endpoints or do we create a service monitor\/collector that will pass on metrics to the user endpoint?
:robot: This issue is automatically synced from: source