Installing RustFS on Linux
Choose a Linux deployment mode — SNSD, SNMD, or MNMD — and install RustFS on one or more servers.
This section covers installing RustFS on Linux servers. For a one-command trial installation, use the Quick Start. For a manual installation, pick one of the three deployment modes below — all three share the same prerequisites and service setup, and differ only in topology and volume configuration.
Single Node Single Disk (SNSD)
One server, one data disk. The simplest mode, with no redundancy — a disk failure means data loss, so rely on backups. Suitable for development, testing, and low-density non-critical business.
→ Single Node Single Disk installation
Single Node Multiple Disk (SNMD)
One server, multiple data disks. Erasure coding shards data across the disks, so the node tolerates a limited number of disk failures, but a whole-server failure still means data loss. Suitable for medium, non-critical business on a single machine.
→ Single Node Multiple Disk installation
Multiple Node Multiple Disk (MNMD)
Four or more servers, each with one or more disks. Erasure coding spans servers, providing disk- and node-level fault tolerance plus horizontal scalability. This is the mode for production workloads.
→ Multiple Node Multiple Disk installation
Before Production
Work through the Pre-Installation Checklists — hardware, network, software, and security — before deploying to production. If you don't need production standards, you can skip them.