HSM/Archiver AGPL for Disk,Tape,Cloud Written in Rust

ticklyjunk1 pts0 comments

HuskHoard — Your Data Doesn't Disappear. It Hibernates.<br>H">

Open Source · AGPL v3 · Built in Rust

Your data doesn't<br>disappear.<br>It hibernates.

HuskHoard is an automated, transparent data-tiering engine for Linux.<br>It keeps your NVMe drives fast by silently archiving cold data to LTO tape,<br>spinning drives, or cloud — while every file stays fully visible to your OS.

Get Started<br>Read the Docs ↗

AGPL-3.0<br>Built with Rust<br>Linux · Kernel 5.1+<br>LTO-5 → LTO-9<br>40+ Cloud Providers

Why HuskHoard

Enterprise storage behavior.<br>Open-source economics.

Enterprise vendors charge thousands to lock your data inside proprietary black boxes.<br>HuskHoard does it for free, in user-space, using formats you already own.

01

[ ⊕ ]<br>Transparent Stubbing

Files stay visible to your OS at full logical size. Applications never know the difference.<br>When a process opens a stubbed file, HuskHoard intercepts via the Linux fanotify kernel API<br>and recalls it in real-time — no FUSE overhead, no mount magic.

Zero FUSE overhead

02

[ ▶ ]<br>StreamGate HTTP Gateway

Watch 4K video directly from LTO tape or S3 via a local HTTP bridge.<br>Plex, Jellyfin, and VLC can seek through massive files instantly using HTTP Range requests —<br>with zero SSD impact and no full recall required.

Built for M&E pipelines

03

[ ◇ ]<br>The Easy Exit Promise

All payload data is stored in standard Zstd streams verified by BLAKE3 .<br>You can extract everything using only dd and zstd.<br>No proprietary formats. No license keys. No asking permission.

No vendor lock-in

04

[ ≡ ]<br>Hardware-Aware by Design

Native SCSI driver for LTO-5 through LTO-9. SMR-safe log-structured writes that eliminate<br>the "write wall." Cloud replication via rclone to 40+ providers, packed into optimal<br>16MB Zstd frames to minimize API costs.

LTO · SMR · S3 · rclone

Architecture

Four components.<br>One seamless system.

Each part of HuskHoard does one job with precision.<br>Together they form a continuous, self-managing storage pipeline.

Component 01

[ DB ]<br>The Catalog

SQLite Brain

Tracks every file, its complete version history, and its exact byte-offset<br>on physical media. The single source of truth for the entire system.

Component 02

[ ◎ ]<br>The Interceptor

fanotify Loop

A lightweight kernel-space listener. Detects when an application requests<br>a stubbed file, blocks the process, triggers instant recall, then resumes.

Component 03

[ ↻ ]<br>The Janitor

Policy Engine

Background daemon that identifies cold data by age, file extension, or<br>directory policy. Queues candidates for the Archive Worker with zero disruption.

Component 04

[ ▤ ]<br>The Archive Worker

Heavy Lifter

Compresses data into seekable Zstd frames, multiplexes writes across the<br>full storage pool, and issues low-level SCSI commands for tape hardware.

Built on the Linux fanotify kernel API,<br>HuskHoard operates entirely in user-space — no kernel modules, no root daemons.<br>N-way replication mirrors cold data simultaneously across local drives, physical tapes,<br>and cloud buckets. Point-in-time recovery lets you roll back any file to a previous<br>version. The BLAKE3 bit-rot scrubber cryptographically verifies offline storage<br>on any schedule you set. Have a look at the Blog for detailed explainers on architecture.

Quick Start

Up and running<br>in minutes.

HuskHoard ships as a single CLI binary. No Docker required. No configuration daemons.<br>Run as your standard user — not root.

huskhoard — bash — Ubuntu 24.04 LTS

# 1. Install system dependencies<br>$ sudo apt install -y build-essential rclone libcap2-bin attr pkg-config libsqlite3-dev git<br>$ curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh && source $HOME/.cargo/env

# 2. Clone and build<br>$ git clone https://github.com/huskhoard/huskhoard.git && cd huskhoard<br>$ cargo build --release<br>Compiling huskhoard v0.1.0 ... Finished release [optimized]

# 3. Grant kernel capabilities (no root daemon needed)<br>$ sudo setcap cap_sys_admin,cap_dac_read_search+ep target/release/huskhoard

# 4. Format your archive volume and start the daemon<br>$ ./target/release/huskhoard format --tape-dev my_archive.img<br>$ ./target/release/huskhoard daemon<br>[HuskHoard] Interceptor active. Janitor online. StreamGate listening :8080

# Stream a file directly from tape — zero disk extraction<br>$ ./target/release/huskhoard cat --file /media/movies/rushes.mp4 | mpv -

Full documentation on GitHub →

Enterprise

Built for scale.<br>Ready for what's next.

The open-source core is free forever under AGPL v3.<br>Enterprise sidecar modules — designed to run alongside the core binary — are on the way.

[ locked ]<br>// DASH<br>Web Dashboard

Real-time "tank gauge" monitoring. Visual capacity planning and tiering activity.

Coming Soon

[ locked ]<br>// METRICS<br>Prometheus Integration

/metrics endpoint ready for Grafana. Full observability for production deployments.

Coming Soon

[ locked ]<br>// REPLICATE<br>Advanced Replication

Fine-grained per-project replication policies. SLA-backed recovery windows.

Coming Soon

[ locked ]<br>// SUPPORT<br>Support &...

huskhoard data file tape kernel release

Related Articles