Introduction
This article is the conclusion to the Chaotic Docker Series, which creates a basis for the PHA Project. Further, it provides an easy way for applying the PHA way for people who do not want to spend a few days to understand Docker and want to directly use it.
Advantages
The goal of Single Source of Information (SSI) is to provide a platform where different software packages can be used modularly in their dockerized forms for different systems. This means that it is a very flexible development platform that would be robust to changes that come during any development process. Another advantage of the SSI is that since this space is shared between the host and the container, no credentials have to be passed to the container. Any changes to the softwares can be made from the host system. Further, if any source installations are made on the SSI, then this does not associate to the size of the container.
Disadvantages
The modularity based on several parameter definitions means the system would be very modular but it also increases the complexity of the project. Most of the modules would be dependant on several parameters that communicate very small details of the system. The biggest rebuttle about this article is that it opens access to several hardware resources of the host system that can make the system vurnerable. For this reason, the project is as of this point only ready for development, it is not production ready. However, with some security based decisions regading the access of the hardware resources, the system can be more secure.
Another point to consider is that if there are several users setup in the host system, the permissions these users have regarding docker must be properly setup. Else the SSI folder might be owned by one user and Docker in general, regardless of which user has access, would be owned by one particular user. This is avoided by setting the host user in the docker container which is documented in
Alternative
One alternative to the docker ros robotics stack concept is IKA ROS and IKA ROS ML. They provide several different versions of ROS Docker and also provide several functions for installations. However, the author would argue (and of course the author tries to sell their idea), that with the SSI these installations could be easily applied to the container. This means that there are no issues with permissions when pulling or pushing with git, but again this project gives a lot of permissions to the containers. Further, if we install with one container, we can also use the software for a container created with the same image. But PHA would not be a consuptions project without look at alternatives without thinking about integrating them, this could be an idea for the future.
TL;DR Docker Compose
This section would be short and sweet. If anyone wants more information, see Docker Compose. Compare between Docker Run Options and Sample Docker Compose to understand the commands used in a compose file.
Introduction
Just remember:
| Command | Definition |
up |
Initialize and start the docker container |
down |
Stop and remove the docker container |
start |
Onyl start an available docker container |
stop |
Only stop a docker container |
SSI, Folder Structure
If you have not created the folder structure as explained in Space for the SSI, setup the folder structure before we create the container:
xhost +local:docker
cd /home/${USER}
mkdir schreibtisch
cd schreibtisch
git clone https://github.com/pradhanshrijal/pha_docker_files
Sample Usage - Compose
Sample up:
docker-compose -f envs/pha-22-mini/docker-compose.yaml up -d
Sample down:
docker-compose -f envs/pha-22-mini/docker-compose.yaml down
For start and stop it’s the same as down.
Sample Usage - Container
With this method we are running docker in the background, now we will look at some additional functions:
Entering Docker:
docker exec -it pha-22-mini /bin/bash
Stopping Docker:
Don’t forget to exit the container if you are inside of it.
docker stop pha-22-mini
Start Docker:
Once you have initialized the container with docker run you don’t have to run the command over and over again, We are storing it for future use:
docker start pha-22-mini
Workspace for Users
Users can install their packages make changes in /home/${USER}/schreibtisch/pha_docker_files/docker_share/git_pkgs. This folder will not disappear if you wish to use docker like an acid bath technique and clear it out for a new fresh installation. It is hight recommended to use a version control system like git to maintain and save your work.
TL;DR PHA Scripts
All these commands can be run with scripts available with PHA Docker. We will see some sample applications:
First of all, go to the main folder we established in the previous section.
cd /home/${USER}/schreibtisch/pha_docker_files
From here we can run simple scripts. Let’s look at the previous example again. First with docker run and then with docker-compose.
Docker Run Scripts
./run.sh -t mini -c pha-22-mini
Don’t for get to stop the container once done.
Compose Scripts
The compose scripts are much better oriented for a multitude of commands.
You can turn a compose script up, with the variables defined in an environement file:
./docker_scripts/run-compose.sh -e docker_scripts/compose-file-mini.env
And even stop, start and turn down an available container:
./docker_scripts/run-compose.sh -e docker_scripts/compose-file-mini.env -o down
There is also a script to directly pass the user to a specified image:
./docker_scripts/run-compose.sh -e docker_scripts/compose-file-user.env -f docker_scripts/docker-compose-user.yaml
Full Usage
Checkout PHA Git Wiki - Usage for the full options available for scripts.
Conclusion
This article series creates a base platform for the PHA Project where different modules of the automated driving stack could be built upon. It gives an extensive guide to using docker and ends with a super short usage guide for docker.
