skills.rest
josefbacik
josefbacikCOMMUNITY
Explore all skills by this author →
GitHub

systing-trace

Capture Linux system traces with the systing BPF tool into DuckDB or Perfetto formats.

174|18|Updated Oct 17, 2024
Search Tags:#performance#debugging#profiling#tracing#linux#bpf
One-click install
npx skills add https://github.com/josefbacik/systing --skill systing-trace
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: systing-trace
Source: https://github.com/josefbacik/systing/tree/main/skills/systing-trace
Command: npx skills add https://github.com/josefbacik/systing --skill systing-trace

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill captures detailed system activity on Linux, including scheduling, stack traces, network events, and syscalls, to diagnose performance issues and understand system behavior.

Core Features & Use Cases

  • Comprehensive Tracing: Records scheduler events, CPU stacks, sleep states, syscalls, network traffic, and Python stacks.
  • Flexible Targeting: Can trace specific PIDs, cgroups, or the entire system.
  • Output Formats: Generates output in DuckDB (recommended for analysis) or Perfetto formats.
  • Use Case: When a web server is experiencing intermittent slowdowns, use this Skill to capture a trace that includes network activity, scheduler events, and syscalls to pinpoint the bottleneck.

Quick Start

Use the systing-trace skill to capture a 30-second system trace to a DuckDB file named 'system_profile.duckdb'.

Frequently Asked Questions about systing-trace

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

FAQPage Schema
How do I trace Linux syscalls and network activity to diagnose performance issues?▼

Linux system tracing captures detailed activity including scheduler events, stack traces, and syscalls using BPF to diagnose performance issues. It outputs the recorded data to DuckDB or Perfetto formats for in-depth analysis of system behavior.

Can I capture BPF traces for specific PIDs or cgroups instead of the whole system?▼

BPF tracing supports flexible targeting, allowing you to capture traces for specific PIDs, cgroups, or the entire system. This focused approach helps pinpoint exact process bottlenecks without generating excessive system-wide noise.

Do I need root privileges to capture system traces on Linux?▼

Capturing Linux system traces with BPF requires root privileges. Elevated permissions are necessary to access kernel-level events, scheduler activity, and CPU stacks for comprehensive performance profiling.

What is the best way to analyze Linux profiling data captured from BPF traces?▼

Outputting the trace to a DuckDB file is recommended for analyzing captured BPF profiling data. DuckDB enables efficient querying of system activity, while Perfetto format is also supported for visualizing scheduling and stack trace events.

What types of system events can I profile with Linux BPF tracing?▼

Linux BPF tracing profiles comprehensive system events including scheduling, CPU stacks, sleep states, syscalls, network traffic, and Python stacks. This detailed capture helps diagnose intermittent slowdowns and understand overall system behavior.

Related Skills

Explore More Software Engineering Skills→
gerald-icagerald-ica

Linux Privilege Escalation

Enumerate Linux systems and exploit vulnerabilities to escalate privileges to root.

Community
Advanced
chrishuffman5chrishuffman5

sqlserver-advisor

Capture SQL Server DMV data into DuckDB and generate prioritized recommendations.

Community
Advanced
antgroupantgroup
16

log-analyzer

Parse and analyze application logs to identify error patterns across services.

Official
Advanced
syxscottsyxscott
1

screen_monitor

Capture screenshots at specified intervals and generate activity summaries.

Community
Intermediate