Skip to content

Helper Script LXC Services

Date: 2026-02-20 System: Proxmox VE 9.1

All LXCs in this document were installed using Proxmox VE Helper Scripts.


LXC 105 - Komodo Core

What it does

A self-hosted Docker management platform used to manage, deploy, and monitor all Docker Compose stacks running in LXC 100. Replaced Dockge (deprecated). Full setup documented in Doc 17.

Installation

Note: As of March 2026, Komodo has been migrated to an addon script in community-scripts. The LXC 105 was created with the original LXC script before this migration.

bash -c "$(curl -fsSL https://raw.githubusercontent.com/community-scripts/ProxmoxVE/main/ct/komodo.sh)"

Updating

After running the one-time migration prompt (y), use:

update_komodo

Details


LXC 106 - Karakeep

What it does

A self-hostable bookmark manager that saves links, notes, and images with AI-based automatic tagging and full-text search. Self-hosted alternative to read-it-later services like Pocket or Instapaper.

AI tagging uses the Nobara PC Ollama (GPU, 192.168.0.100:11434) - requires Nobara to be on. See Doc 13 for setup details.

Installation

bash -c "$(curl -fsSL https://raw.githubusercontent.com/community-scripts/ProxmoxVE/main/ct/karakeep.sh)"

Details

  • Port: 3000
  • LXC ID: 106

LXC 107 - n8n

What it does

A workflow automation platform for building automated pipelines connecting apps and services.

Installation

bash -c "$(curl -fsSL https://raw.githubusercontent.com/community-scripts/ProxmoxVE/main/ct/n8n.sh)"

Details

  • Port: 5678
  • LXC ID: 107

LXC 108 - Ollama (decommissioned)

Decommissioned 2026-07-21: LXC 108 was unused (Karakeep, n8n, and all docker-host stacks used the Nobara PC Ollama instance instead) and was deleted to free space on the pve/data thin pool. Section kept for reference; see hosts/ollama.md (moved to deprecated/) for full history.

What it did

Ran open-source large language models locally. This instance was always-on, CPU-based, and intended for light inference tasks. In practice nothing referenced it - all AI tagging (Karakeep) and agent workloads (Hermes/Odysseus) use the Nobara PC Ollama instance (GPU, 192.168.0.100:11434) instead.

Installation (historical)

bash -c "$(curl -fsSL https://raw.githubusercontent.com/community-scripts/ProxmoxVE/main/ct/ollama.sh)"

Details (historical)

  • Port: 11434
  • LXC ID: 108
  • OLLAMA_HOST: set to 0.0.0.0 to allow access from other LXCs