troubleshoot-cassandra

What problem does it solve?

It helps diagnose Apache Cassandra production incidents by turning confusing symptoms (GC spiral, compaction spiral, tombstone storm, disk exhaustion, hint overflow) into a structured triage path with concrete Netdata signals.

Core Features & Use Cases

• Guided failure-pattern triage: Maps observed behavior to dominant Cassandra failure archetypes and orders investigation by the operator playbook’s signal domains.
• Netdata-native evidence via MCP: Uses MCP queries to validate node liveness, transport health, request rate/latency, and related signals from the Netdata Cassandra collector.
• Actionable remediation verification: Recommends remediation pointers and verifies the fix by re-checking the same load-bearing contexts to confirm signals return to expected bands.

Use case: An on-call engineer is paged for elevated Cassandra errors and latency, and needs to quickly determine whether the incident matches GC pressure, compaction backlog, tombstone amplification, disk space constraints, or hint backlog inconsistencies.

Quick Start

Ask: "Diagnose Apache Cassandra issues for host <hostname> with Netdata MCP and determine whether it is a GC death spiral, compaction death spiral, tombstone storm, disk space exhaustion, or hint overflow, then verify the likely fix using the required load-bearing metrics."