Define alert criteria for F3 metrics

[masih]

We need to define the criterion on which an alert should be triggered, categorised by criticality. This is needed by node operators, exchanges and anyone running f3 in production environment with SLAs attached.

Tasks

Beta Give feedback

1. Document the alert criterias
2. Define them in grafana/add them to grafana dashboards
3. Write runbook for each of the alerts, i.e. what one should do when alert triggers

[masih]

Also see:

• https://www.notion.so/filecoindev/F3-Launch-5cdce4f1aa6e4c398b475f6e690c47fe#66c2deefcd854376a453d5f84ebf4a5b
• Report command semantics on slack, in this section: https://www.notion.so/filecoindev/F3-Launch-5cdce4f1aa6e4c398b475f6e690c47fe?pvs=4#fcbaa14d8eab4d548eeba5c3545e8dfd

[rjan90]

Can be closed once we have converted the runbook to actual alerts, and have the integration to fil-lotus-alerts up and running,