Prerequisites & Server Requirements¶
Minimum system specifications¶
| Native / PM2 | Docker | |
|---|---|---|
| OS | Ubuntu 20.04/22.04 LTS or Debian 11/12 (recommended); RHEL/AlmaLinux supported | Ubuntu 22.04 LTS, RHEL 9, Debian 12, or Amazon Linux 2023 |
| CPU | 2 vCPUs minimum (4 recommended) | 4 vCPUs (x86_64 or ARM64) |
| RAM | 4 GB minimum (8 GB recommended) | 8 GB |
| Disk | 20 GB free SSD | 40 GB SSD |
| Network | Inbound HTTP (80) or HTTPS (443) | Inbound HTTP (80) or HTTPS (443) |
Software prerequisites¶
Installed automatically by install-onprem.sh if not already present: Node.js 20, PM2, PostgreSQL, Redis, and Nginx.
Install yourself before starting:
- Docker Engine 24.0+
- Docker Compose plugin 2.20+
- OpenSSL (for license verification)
docker --version
docker compose version
What you'll receive from your vendor¶
- A distribution archive:
itsm-onprem-v<version>.tar.gz(compiled, source-free) - A
license.keyfile (RSA-signed, tied to your seat count and feature tier)
AI features (optional, both deployment types)¶
Incident summarization, ticket triage, RCA drafting, and SLA risk prediction run in a separate, stateless AI microservice (itsm-ai). It's never exposed to the internet — the backend talks to it internally. If it's stopped, the rest of the app keeps working; only AI-powered suggestions stop. Configure it via LLM_PROVIDER and OPENAI_API_KEY/GEMINI_API_KEY in .env.