Browse documentation
Understanding redundancy-aware impact
Not every failure is equally serious. If one of two redundant uplinks fails, service continues — but if the only uplink fails, everything behind it goes down. JOATOS's redundancy-aware impact understands this by combining live status with your dependency map.
How impact is classified
For each entity that depends on others, JOATOS works out live impact:
- Full — the entity is down, or all of its required providers are down, or every member of a redundancy group is down.
- Partial (degraded) — some but not all members of a redundancy group are down, so it's running reduced.
- Unaffected — a required provider is up, or a redundancy group still has a healthy member.
Redundancy groups
The magic is the redundancy group. When you create a dependency, you can tag it with a redundancy group. Edges that share a group on the same dependent are treated as mutually redundant — any one being up satisfies the group. Ungrouped dependencies are treated as single points of failure.
So two uplinks in the same redundancy group = degraded if one fails; an ungrouped single uplink = full outage if it fails.
The live impact panel
The Live dependency impact panel on the Status page shows this in real time: a summary line (down · affected (full) · degraded (redundancy)) and a list of affected entities, each with a Full or Partial badge and a plain-language reason such as "1 of 2 redundant uplinks down". Each links to its detail page.
Notes & tips
- This view is read-only — it shows current impact but doesn't create outages on its own. Report an outage to track an incident.
- The quality of the impact view depends on your dependency map — record dependencies and tag redundant paths so JOATOS can tell degraded from down.