User Story:

As an ACM user I want to display abbreviated information on health of my cluster fleet and get faster to actionable events. I want to leverage Red Hat experience in supporting fleet of OpenShift cluster and consume same information as Red Hat internal support/SRE team. I'm specifically interested in understanding the connections between events and consequences of these to the cluster (fleet) health.  

Goal:

Insights TDI team exposes Insights analytics work in internal dashboards to internal teams (see CCX-238) . This includes information about Upgrade risk, grouping of cluster events and correlations between them. To overall goal of this epic is to expose Insights TDI work to external customers within ACM. 

Review FleetView mocks/internal dashboard, produce design and architecture to integrate with ACM. Focus on adding unique features to ACM that are result of internal analytics work and resutls of TDI team. 

Following information will be presented to ACM admin. Information will be available for all OCP customers providing Insights data (connected to consoledot)

• OCP Version & Upgrade success rate (Upgrade risk - see CCX-216 )
• Fleet Operator conditions & trends
• Fleet Insights recommendations and trends
• Fleet Resource usage, eg, consumption and trends
• Fleet Alerts overview including trends and most significant alerts (Alerts noise reduction - TDI-38)
• Integrate key information from Alerts, Recommendations, Upgrade risk and Operator conditions into the ACM Overview page
• Define the UX and deliver management of the Alerts (and other events), for example silencing groups of Alerts, sorting Alerts via Label, and viewing Alerts over timeline (history) 
• Any other updates from Fleet Insights integration work - CCX-238 

Releases:

• v1.0 - provide dashboard of information available in ACM's Thanos, identify missing pieces that require consoledot API endpoints
• v2.0 - implement the new APIs, expand the ACM visualizations using these APIs
• v3.0 - analytical features (v2 is likely to only include a subset of capabilities already available in our internal pipelines, i.e. raw data and its aggregation, no advanced analytics)

Mockups:

• https://docs.google.com/presentation/d/1QB3W1-3dTJPfH30wD-NhSAKswTh6EjYAcyLLf6iuqY4/edit#slide=id.g20ffe799602_0_10 
• Dashboard - https://superset.datahub.redhat.com/superset/dashboard/ccx-fleet-insights/ 

Requested by: sberens@redhat.com 

Architects: jbanerje@redhat.com , inecas@redhat.com 

UXD: kriker , smckinno