Introduction
TrixOS is a self-hosted panel for managing dedicated servers. It talks to every server’s BMC over IPMI or Redfish, installs and rescues servers over PXE, keeps their hardware inventory and disk health, manages IP addresses and switch ports, and measures bandwidth. Customers and resellers get their own logins and API tokens, and an API that is compatible with TenantOS lets an existing control panel keep working.
These docs are for the people who install and run TrixOS: administrators of a hosting company, not their customers.
How TrixOS is built
Section titled “How TrixOS is built”| Part | What it does |
|---|---|
| Panel (control server) | Web UI, API, job worker, poller and boot-file cache, run with Docker Compose on one Linux host, with Postgres and Redis. |
| Site agent | A small service on a Linux host in each datacenter. It joins the panel over WireGuard, routes the panel into the out-of-band (BMC) network and serves DHCP and TFTP to servers that have an install job running. |
| BMCs | iLO, iDRAC, XClarity, Supermicro or any IPMI/Redfish BMC. TrixOS never needs an agent on the servers themselves. |
Everything that changes real hardware runs as a job: a sequence of steps with a live log, one job per server at a time. Power actions, OS installs, rescue systems, disk wipes and password resets are all jobs.
Where to start
Section titled “Where to start”The full API is documented at api.trixos.sh, and every release is listed in the changelog.