Docker image for Duniter

Building

Build this image:

docker build . -t duniter/duniter -f release/docker/Dockerfile

Usage

Start the node:

docker run -d -p127.0.0.1:9220:9220 -p10901:10901 -p20901:20901 --name duniter duniter/duniter

You can execute duniter commands on the started container, for example:

docker exec -it duniter duniter wizard key
docker exec -it duniter duniter sync g1.duniter.fr 443

You also can open a new terminal on the started container with:

docker exec -it duniter /bin/sh

The web UI can be accessed from the host machine by browsing http://localhost:9220.

Stop the node:

docker stop duniter

Restart the stopped node:

docker start duniter

Features

Volumes

The container creates 2 docker volumes. The first one is mounted under the Duniter home directory, i.e. /var/lib/duniter. This is the place the Duniter configuration files and database will be stored. The other one is mounted under /etc/duniter which is where it is recommended to store the keyfile when the instance is a member node, so that the keyfile is kept separated from the - potentially - volatile data.

See more about managing volumes with docker in the docker documentation.

If you mount one or more of these directories on your host, be careful that the files must be accessable by the in-image duniter:duniter user (uid:gid is 1111:1111).

Keyfile

If the /etc/duniter/keys.yml is found on the image, it will automatically be used as keyfile for the server. If you have such a file in your current directory, you can start the node with:

docker run -d -p127.0.0.1:9220:9220 -p10901:10901 -p20901:20901 --mount src="${PWD}",dst=/etc/duniter --name duniter duniter/duniter

Note that your file must be readable by the in-image duniter:duniter user (uid:gid is 1111:1111).

Image parameters

If you give parameters to the image when creating container, they will be given to the Duniter node. In this case, it will not start the Duniter daemon. Example:

docker run -it --name duniter duniter/duniter logs

When no parameters are given, duniter is called with the command direct_webstart.

Note that you should not call duniter with daemon command (webstart or start) if you run docker with -d parameter, because the docker image will then immediately stop.

Environment variables

To ease the deployment automation, three environment variables are available:

• DUNITER_MANUAL_CONFIG (boolean, default = false)

  When set to 'true' (or 'yes', or '1'), the entrypoint script waits until the file '/etc/duniter/conf.json.orig' is present before starting the duniter service. Here is the workflow when enabled:

  1. wait for '/etc/duniter/conf.json.orig'
  2. if this file was changed since the previous startup:
     1. Save 'conf.json' to 'conf.json.old'
     2. Replace 'conf.json' with 'conf.json.orig'
     3. Save the new file's checksum to compare with at next startup
  3. continue the startup sequence

  When set to false, the startup sequence continues.

• DUNITER_AUTO_SYNC (boolean, default = false) Requires 'DUNITER_SYNC_HOST'.

  When set to 'true' (or 'yes', or '1') and the folder '/var/lib/duniter/duniter_default/data' doesn't exist, a 'duniter sync' command will be issued before starting the service.

  When set to false, the service is started directly.

• DUNITER_SYNC_HOST (hostname, default = "")

  This is the 'host:port' parameter to use with 'duniter sync' when 'DUNITER_AUTO_SYNC' is enabled.

  The synchronization won't be launched when the variable is not defined or empty.

Test development version

To test development version on G1-test network:

docker run -d -p127.0.0.1:9330:9220 -p10900:10900 -p20900:20900 -v $HOME/duniter-docker-home:/var/lib/duniter --name registry.duniter.org nodes/typescript/duniter:dev sync g1-test.duniter.org