Blast radius
Clusters, components and the calls between them, shaded by impact. Drag the timeline to replay how the failure spread. Select a component to see its evidence.
Probable root causes
Ranked by how strongly the evidence supports each one. Select any piece of evidence to see the exact line you pasted.
Rollback options
Ways to stop the damage, fastest safe option first, with the exact commands for your clusters.
Error budget
How much of the service level objective (SLO) error budget this incident is using, and how fast — using the burn-rate method from the Google Site Reliability Engineering (SRE) Workbook.
Investigate with Claude
Optional. A second opinion on the same evidence, kept read-only.
Observability stacks
How four common setups would handle this incident at your scale, and where each would strain. Every figure links to its source; anything not sourced is marked as reasoning.