claude-memory/graph/fixes/high-cpu-on-docker-host-from-avahi-daemon-and-gnome-775506.md
Cal Corum b140d4d82a migrate: 313 memories from MemoryGraph
- 313 new markdown files created
- 30 relationships embedded
- 313 entries indexed
- State initialized with usage data
2026-02-13 11:11:48 -06:00

41 lines
1.2 KiB
Markdown

---
id: 775506ce-ed06-4c36-95f3-73379855b44a
type: fix
title: "High CPU on Docker host from avahi-daemon and GNOME"
tags: [cpu, docker, avahi, gnome, gdm, performance, homelab, fix]
importance: 0.7
confidence: 0.8
created: "2025-12-20T04:36:05.718983+00:00"
updated: "2025-12-20T04:36:05.718983+00:00"
relations:
- target: 62ee21e8-2b56-4d38-a73d-47e2724f08c6
type: BUILDS_ON
direction: incoming
strength: 0.5
---
**Problem:** Docker host showing constant ~12% CPU usage from background services.
**Root Cause:**
1. avahi-daemon (mDNS) constantly processing Docker veth interface changes (67+ hours accumulated CPU)
2. gvfs-udisks2-volume-monitor (GNOME disk monitor) running unnecessarily on server
3. gdm and gnome-shell running on headless server
**Solution:**
```bash
# Disable avahi (including socket to prevent respawn)
sudo systemctl stop avahi-daemon.socket avahi-daemon
sudo systemctl disable avahi-daemon.socket avahi-daemon
# Disable GNOME desktop
sudo systemctl stop gdm
sudo systemctl disable gdm
# Kill lingering processes
sudo pkill -f gvfs-udisks2-volume-monitor
```
**Result:** ~12% CPU and 180MB RAM recovered.
**When to apply:** Any Docker host with desktop environment installed that doesn't need GUI access.