cpu-saturation

Attribute high CPU saturation on StarRocks FE or BE nodes to process, module, and workload.

73|12|Updated Apr 24, 2026
One-click install
npx skills add https://github.com/StarRocks/starrocks-debug-skills --skill cpu-saturation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cpu-saturation
Source: https://github.com/StarRocks/starrocks-debug-skills/tree/main/cpu-saturation
Command: npx skills add https://github.com/StarRocks/starrocks-debug-skills --skill cpu-saturation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps diagnose StarRocks CPU saturation by identifying which process, module, and workload are consuming the most CPU on FE or BE hosts, so teams can stop guessing and start the right investigation.

Core Features & Use Cases

  • CPU Attribution: Separates query execution, import flush and compression, compaction, clone tasks, FE planning or GC, and non-StarRocks processes.
  • Multi-layer Triage: Uses OS signals, BE CPU task metrics, process and thread inspection, and hot-function profiling to narrow the cause quickly.
  • Directed Handoff: Routes incidents to the correct deep-dive path for query, import, compaction, or node-level troubleshooting.
  • Use Case: When a production cluster shows high load average and users report slow queries, this Skill determines whether the bottleneck is a runaway query, background compaction, an import burst, or an external process.

Quick Start

Use the cpu-saturation skill to identify whether StarRocks CPU pressure comes from queries, imports, compaction, FE GC, or another process, and then route the incident to the right deep-dive skill.

Frequently Asked Questions about cpu-saturation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find out what is causing high CPU saturation on StarRocks FE or BE nodes?

To diagnose high CPU saturation on StarRocks nodes, inspect OS processes, BE CPU task metrics, and perf or pstack hot-functions to attribute the load to a specific process, module, and workload.

What causes high CPU load during StarRocks compaction or import flush tasks?

High CPU load during StarRocks compaction or import flush tasks is caused by background compression and clone tasks. Profiling BE CPU task metrics and OS processes isolates the specific module driving the workload.

How do I use perf top or pstack to troubleshoot StarRocks query slowdowns?

Use perf top or pstack to analyze hot-functions and determine if StarRocks query slowdowns stem from query execution, FE planning, FE GC, or background tasks causing CPU saturation.

Does this CPU triage process separate StarRocks internal workloads from external processes?

Yes, this CPU triage process separates StarRocks internal workloads like queries, imports, and compaction from external non-StarRocks processes by inspecting OS process and thread CPU consumption.

What is the best way to triage a StarRocks production cluster with a high load average and slow queries?

The best way to triage a StarRocks cluster with high load average and slow queries is multi-layer CPU attribution, routing the incident to query, import, compaction, or node deep-dive paths based on OS and BE metrics.