Longhorn Engine is a lightweight block device storage controller that implements network RAID functionality for distributed storage.
The tool solves the problem of providing reliable block storage across multiple nodes by implementing a controller that synchronously replicates write operations to multiple replicas backed by Linux sparse files. It presents replicas as networked disks supporting read/write operations over a network protocol, with a kernel driver frontend that translates block device operations on a mapped device to user-level network requests. The controller detects faulty replicas and automatically rebuilds them, while coordinating snapshot and backup operations. Each block device is backed by its own dedicated controller, and both controllers and replicas run as Docker containers.
Longhorn Engine suits projects requiring distributed block storage with built-in redundancy and snapshot capabilities. It is particularly relevant for containerized environments where storage needs to be replicated across multiple hosts. The tool supports Open-iSCSI and tgt frontends for kernel-level block device access. Users should be aware that setup involves configuring network connectivity between replica and controller containers, and that the `longhorn` command provides management operations including replica management, snapshot creation, and backup coordination.
The project maintains active continuous integration through automated build workflows. Development activity shows consistent engagement with the codebase through regular updates and maintenance of the core storage controller functionality.
