This is a Docker implementation for Edomi, a PHP-based smarthome framework.
For more information please refer to Official website or Support forum
You have two options:
- Pull the ready-made image from DockerHub or
- Build from scratch
The Edomi archive together with all required packages will be downloaded during docker build. I've added openssh-server and additionally I've set the root password to '123456'.
sudo docker pull starwarsfan/edomi-docker:amd64-latestThe image build is split into two separate build steps. The first step generates updated CentOS base image with all required packages. The second step build the Edomi image, which is based on the image from the first build step.
sudo git clone https://github.com/starwarsfan/edomi-baseimage.git
sudo git clone https://github.com/starwarsfan/edomi-docker.gitcd edomi-baseimage
sudo docker build \
-f amd64.Dockerfile \
-t starwarsfan/edomi-baseimage:amd64-latest .If you built an own Edomi-Baseimage (the step before), you need to update it's reference
on the first line of the Dockerfile. The default first line looks like the following snippet,
where the version should be replaced with latest, if you build your personal base image
with latest as version:
FROM starwarsfan/edomi-baseimage:amd64-latestAfterwards you can trigger the build with:
cd edomi-docker
sudo docker build \
-f amd64.Dockerfile \
-t starwarsfan/edomi-docker:amd64-latest .You can pass a different root passwort to the build and you can pass the Edomi version to download too:
sudo docker build \
-t starwarsfan/edomi-docker:amd64-latest \
--build-arg ROOT_PASS=Th3Passw0rd \
--build-arg EDOMI_VERSION=EDOMI-Beta_156.zip .sudo docker run \
--name edomi \
--restart=on-failure \
-p 80:80 \
-p 8080:8080 \
-p 3671:3671/udp \
-p 50000:50000/udp \
-p 50001:50001/udp \
-p 22222:22 \
-e KNXGATEWAY=192.168.178.4 \
-e KNXACTIVE=true \
-e WEBSOCKETPORT=8080 \
-e HOSTIP=192.168.178.3 \
-d \
starwarsfan/edomi-docker:amd64-latestWith this configuration the edomi web instance is reachable via URL http://<docker-host-ip>/admin or https://<docker-host-ip>/admin and the commandline via ssh with ssh -p 22222 <docker-host-ip>. With the (optional) parameters KNXGATEWAY, KNXACTIVE and HOSTIP you can pre-configure some settings for Edomi. Leave it empty to do this via the Edomi admin webpage. Keep in mind to set "global_serverIP" in Edomi (or via docker run parameter 'HOSTIP') to your Docker host IP. Otherwise the KNX communication probably will not work. Change http and/or https port to your needs.
If you use other Edomi modules which communicate using dedicated ports, you need to map them using additional -p : parameters.
It is important to map all used ports! According to the example with the default values above, here's a short description:
-
-p 80:80
Mapping of used http port to Edomi http port.
-
-p 8080:8080
Mapping of Websocket port. These values must be the same on both sides of the colon and correspond to the configuration value on Edomi base configuration. The "WEBSOCKETPORT" variable will allow to set the websocket port in the edomi configuration. All three values (both sides of the mapping as well as the "WEBSOCKETPORT" value) must match.
-
-p 3671:3671/udp
Mapping of used port for KNX traffic. As this is UDP traffic, it must be finished with "/udp" right after the internal port, which must correspond to the configuration value on Edomi base configuration.
-
-p 50000:50000/udp
Mapping of used port for KNX control endpoint. As this is UDP traffic, it must be finished with "/udp" right after the internal port, which must correspond to the configuration value on Edomi base configuration.
-
-p 50001:50001/udp
Mapping of used port for KNX data endpoint. As this is UDP traffic, it must be finished with "/udp" right after the internal port, which must correspond to the configuration value on Edomi base configuration.
-
-p 22222:22
Mapping of used ssh port to access the container using ssh.
Please note:
It is important to use the option --restart=on-failure because it is used to handle Edomi shutdown or restart from the admin ui. The trick is to exit the container with a non zero exit code in case Edomi should be restartet. If it should be shut down, the exit code will be zero, which is not a failure for Docker and so the container will not be restartet again.
The image offers three mountpoints:
- /var/edomi-backups
- /var/lib/mysql
- /usr/local/edomi
So it is possible to use dedicated volumes, which enables the possibility to reuse the volumes on a new container.
With the additional run parameter -v : or -v : you can mount a folder from the Docker host or a Docker volume into the container.
The usage of volumes should be preferred, as this offers the most flexibility. To do so, at first you should create empty volumes:
sudo docker volume create edomi-backups
sudo docker volume create edomi-db
sudo docker volume create edomi-installationNow the container can be started using these volumes:
sudo docker run \
--name edomi \
-v edomi-backups:/var/edomi-backups \
-v edomi-db:/var/lib/mysql \
-v edomi-installation:/usr/local/edomi \
...If a new container is created using empty volumes, then the content which is already existing on the used location inside the used Docker image is copied onto the volume. So if you docker run a new Edomi instance, the whole content from these three mountpoint directories will be copied to the used volumes. So if the container instance is destroyed and the volumes where used on a new Edomi instance, the content from the previous instance will be there.
The usage of a directory from the Docker host instead of a volume is similar in it's usage. You need to use the full path to the folder on the left side of the colon:
sudo docker run \
--name edomi \
-v /data/edomi-backups/:/var/edomi-backups/ \
...With the additional run parameter -v /:/ you can mount a file on the docker host into the Edomi container. This is useful if you use LBS like LBS19000690 (Jahrestage), which require access to separate files. So the run command may look like the following example:
sudo docker run --name edomi -v /home/edomi/feiertage.csv:/usr/local/edomi/www/visu/feiertage.csv ...- Backup current Edomi instance using
Verwaltung > Datensicherung > Backup herunterladen - Create volumes to store data, see 3.1 above
- Start new container using created volumes
- Copy downloaded backup to created backup-volume
- Import backup using
Verwaltung > Datensicherung > Wiederherstellung
This instruction works for a Centos7 docker host. Other distributions may need some adjustments.
sudo tee /etc/yum.repos.d/docker.repo <<-'EOF'
[dockerrepo]
name=Docker Repository
baseurl=https://yum.dockerproject.org/repo/main/centos/7/
enabled=1
gpgcheck=1
gpgkey=https://yum.dockerproject.org/gpg
EOFsudo yum install docker-engine -ysudo systemctl enable docker.servicesudo systemctl start docker.serviceCheck running / stopped container:
sudo docker ps -aStop the container
sudo docker stop edomiStart the container
sudo docker start edomiGet logs from container
sudo docker logs -f edomiOpen cmdline inside of container
sudo docker exec -i -t edomi /bin/bashThe following steps can be used if you're facing issues during the update from version 1.62 to 1.63.
To fix the update problems between these two versions, the image contains a fixed version of the file, which causes the update failure. Here are the steps to do:
- Pause (not stop!) Edomi using the Admin-UI
- Execute the following cmd:
$ docker exec -it <name-of-your-container> /root/patch_1.62.sh - Update Edomi to 1.63 using the Admin-UI
These variant could be used if you're running the Edomi container on a management tool like Portainer. To do so, perform these steps:
- Pause (not stop!) Edomi using the Edomi Admin-UI
- Open the cmdline on the container
- Execute the following cmd:
$ /root/patch_1.62.sh - Update Edomi to 1.63 using the Admin-UI