A simple Docker Compose setup to manage running hydrus server, hydrusrv and hydrusrv-sync
Adandoned: hydrusrv-docker is no longer maintained.
This is a Docker Compose setup to run hydrus server, hydrusrv and hydrusrv-sync together. It shares a volume between the three containers that contains the databases and files of hydrus server as well as the databasees of hydrusrv.
The easiest way to install is via cloning this repository:
user@local:~$ git clone https://github.com/mserajnik/hydrusrv-docker.git- Docker, including Docker Compose
If you have installed via cloning the repository, you can update via Git:
user@local:hydrusrv-docker$ git pullAfter updating, compare your docker-compose.yml with
docker-compose.yml.example in case that file has changed with the update and
adjust your configuration accordingly. Most of the time, updates will bump the
version of one or more of the services.
hydrusrv-docker follows semantic versioning and any
breaking changes that require additional attention will be released under a new
major version (e.g., 2.0.0). Minor version updates (e.g., 1.1.0 or 1.2.0)
are therefore always safe to simply install via the routine mentioned before.
When necessary, this section will be expanded with upgrade guides to new major versions.
With 2.0.0, compatibility for newer hydrusrv versions (4.x+) and
hydrusrv-sync has been added. Since hydrus server,
hydrusrv and hydrusrv-sync are now
all available on Docker Hub, hydrusrv-docker can just use those
images instead of forcing the user to build them himself.
All three containers can be configured fully via docker-compose.yml (no more
editing hydrusrv's .env file) and those changes apply on container creation
instead of on image build which makes it a lot faster and more convenient to
adjust settings. Therefore, you need to replicate the settings you would have
set via hydrusrv's .env before in docker-compose.yml and also configure
hydrusrv-sync.
Instead of using a bind mount, the recommended way to persist data is now via
named volume. The easiest way to move your existing data to the volume is to
spin up the containers and shut them down again – this will create the named
volume on the first startup. Next, go to the volume directory on the host
machine (usually /var/lib/docker/volumes/<volume name>) and replace all the
data inside with your existing contents in the data directory.
Lastly, rename your application database (usually app.db) to
authentication.db to finish the migration (or simply delete app.db if you
do not care about losing any existing users/tokens – in that case, hydrusrv
will create a new authentication.db itself on the next startup).
Duplicate docker-compose.yml.example as docker-compose.yml:
user@local:hydrusrv-docker$ cp docker-compose.yml.example docker-compose.ymlFirst, you have to set the volume used by the containers. Take a look at the
volumes section in each service block. It is recommended to stick to the
default setup (using a named volume) for performance and ease of use. If you
decide to do so and are happy with the volume name (hydrusrv-docker-data),
you do not have to change anything for this step.
You can also use a bind mount instead of a volume if you wish. Whatever you do,
make sure to use the mount point /data inside the containers to avoid issues
and do not forget to also adjust the top-level volumes block (outside the
three service blocks) if you rename the default volume.
Next, take a look at the settings for hydrusrv and
hydrusrv-sync and adjust the environment
sections of both services as needed. The paths for database files, access log
file and hydrus server media files are already set up to work with the default
volume setup. If you change the PORT environment variable of hydrusrv, make
sure to also adjust its ports section accordingly.
Finally, the hydrusrv-sync container will run syncs in intervals using a cron
job. By default, the schedule is 0 4 * * *, which translates to every day at
4am. If you wish to change this schedule, simply adjust the
DOCKER_CRON_SCHEDULE environment variable in the hydrusrv-sync service block.
You can spin up the containers like this:
user@local:hydrusrv-docker$ docker-compose up -dTo check if they have not crashed when starting (this would indicate a configuration error), you can use the following command:
user@local:hydrusrv-docker$ docker psThis should show three running containers (unless you are running additional ones); one for hydrus server, one for hydrusrv and one for hydrusrv-sync.
Alternatively, you can also open a web browser and hit https:/localhost:45870
for hydrus (needs to be https; your browser will likely show a certificate
issue, but that is because hydrus uses a self-signed one – you can safely
ignore that warning) and http://localhost:8000/api for hydrusrv to see if
they are working correctly.
To stop the containers and remove them, use the following command:
user@local:hydrusrv-docker$ docker-compose downYou will have to do this every time you update hydrusrv-docker. Simply running
docker-compose up -d again afterwards will re-create the containers from the
updated images.
From time to time, you will want to remove old and no longer used images and containers. To do this, you can use the following command:
user@local:hydrusrv-docker$ docker system pruneI recommend doing this after updating and re-creating the containers so it will automatically remove anything that is no longer in use.
For further information, take a look at the Docker Compose documentation.
You are welcome to help out!
Open an issue or submit a pull request.
MIT © Michael Serajnik