https://github.com/TrilliumIT/docker-zfs-plugin

Docker volume plugin for creating persistent volumes as a dedicated zfs dataset.
https://github.com/TrilliumIT/docker-zfs-plugin

docker docker-volume-plugin docker-volumes zfs zfsonlinux

Last synced: 3 months ago
JSON representation

Docker volume plugin for creating persistent volumes as a dedicated zfs dataset.

• Host: GitHub
• URL: https://github.com/TrilliumIT/docker-zfs-plugin
• Owner: TrilliumIT
• License: mit
• Created: 2016-08-18T19:43:26.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2023-02-24T23:59:50.000Z (almost 2 years ago)
• Last Synced: 2024-08-01T12:23:52.419Z (6 months ago)
• Topics: docker, docker-volume-plugin, docker-volumes, zfs, zfsonlinux
• Language: Go
• Size: 43 KB
• Stars: 91
• Watchers: 8
• Forks: 13
• Open Issues: 10
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

• awesome-starred - TrilliumIT/docker-zfs-plugin - Docker volume plugin for creating persistent volumes as a dedicated zfs dataset. (docker)

README

        
# docker-zfs-plugin

Docker volume plugin for creating persistent volumes as a dedicated zfs dataset.

# Installation

Download the latest binary from github releases and place in `/usr/local/bin/`.

If using a systemd based distribution, copy

[docker-zfs-plugin.service](docker-zfs-plugin.service) to `/etc/systemd/system`.

Then enable and start the service with `systemctl daemon-reload && systemctl

enable docker-zfs-plugin.service && systemctl start docker-zfs-plugin.service`.

* Usage

After the plugin is running, you can interact with it through normal `docker volume` commands.

Recently, support was added for passing in ZFS attributes from the `docker volume create` command:

`docker volume create -d zfs -o compression=lz4 -o dedup=on --name=tank/docker-volumes/data`

* Legacy

The driver was refactored to allow multiple pools and fully qualified dataset names. The master branch has removed all legacy naming options and now fully qualified dataset names are required. If you still have not converted to fully qualified names, please use the latest release in the v0.4.x line until you can switch to non-legacy volume names.