Spike: Identify the most active app/service/component that would benefit from monitoring

[hemajv]

Identify the most active app/service/component that can be used as an example for outlining the end-end monitoring workflow from setting up operation tooling, collecting and storing metrics, interacting with metrics (notebooks etc) and visualizing the metrics.

[pacospace]

Some initial list

• Kebechet
• Meteor
• JupyterHub
• PostgreSQL database
• pipelines with argo workflows
• pipelines with Tekton

[hemajv]

Initial doc outlining the monitoring workflow: https://docs.google.com/document/d/1xAkyLDXgAG7V_4u6NY28L93COMPzsHG7Xs69JUV8aVQ/edit?usp=sharing

[hemajv]

I think we can consider Kebechet as our example application as it would be a good use case to explore how we monitor an application consisting of multiple dependent components, monitoring data sources etc.

thoughts @harshad16 @pacospace @xtuchyna @goern @durandom?

[goern]

thoth-station/kebechet#546