Deploy Python service with

This guide helps you to deploy a Python service in a Docker container using the D2C platform. It might be helpful for beginners as well as for advanced developers.

Denis Zaripov

Product manager

Why Docker?

A container image is a lightweight, stand-alone, executable package of a piece of software that includes everything needed to run it: code, runtime, system tools, system libraries, settings. Containers isolate software from its surroundings, for example, differences between development and staging environments and help reduce conflicts between teams running different software on the same infrastructure.

Please, check out the original Docker post about containers for a better understanding of all the benefits of the technology.


It’s easier to start if you have already connected a host.

You can start with our Demo host, but it is created only for one hour. If you need more time, please follow our guides for connecting cloud providers or connecting your own machine to D2C.

Creating Python service

At first, you need to open or create any project and click “+Create service“. You will see a list of services which can be deployed with D2C:

 Let’s go ahead and click on the Python bar.

Creating Python service. Service settings


The name should start with a letter and contain up to 16 characters (Latin letters, numbers, hyphen).

Each service has its unique name. Services can communicate with each other by container names (e.g. shadow-1) or alias-names (e.g. shadow). Moreover, we use them to create public domains like: [servicename]-[www].[accountID].[at] for your services which are served by NGINX or HAProxy. We will talk about edge services in the next articles.


You can choose a version for your application from a list:

If your application requires node.js or its modules (e.g. for building) you can choose what you need to install into a container.
Global dependencies

In this field, you can specify commands for installing global dependencies and modifying Docker image of your service. For modifying source code use Local dependencies.

Examples: pip install, bundle install, apt-get install, npm install -g

Creating Python service. Application source


You can choose what to use for getting application sources. The most recommended is Git. If you use a private repository you should add an SSH key to your account. Here are manuals about adding SSH keys into your GitHubBitBucketGitLab accounts. Another way is to use Login/Password, but the best practice is to use SSH keys. 


Another method is to specify a link to sources. 

Protocols: http, https, ftp (for closed ftp you should specify login/password).
File formats: .tar.bz2, .tar.gz, .tar, .zip


Moreover, you can upload from your machine.

File formats:

Maximum size: 50MB

Creating a Python service. Ports 

This is an important part.

By default, application containers are started inside a private network and have dynamically assigned local IP addresses. Apps can reference each other by container name. It does not matter on which host the app is running – all private network intercommunication is transparent for all services in your account.

Access from the Internet is disabled by default (except edge services). You can enable access from the Internet while creating or editing service. For example, if you publish your application on port 8000, you can access it at ip_address_of_a_host:8000

Creating Python service. Application source

Local dependencies and code’s preparation
Commands for installing local dependencies and making your code ready to work. Examples: npm install, composer install, bower install, etc. or do some for preparation: gulp build, grunt build, etc.
Start command

Start command for your application

Initial commands
Commands which are executed only once on the first container after the first deploying a service. You can use it for populate database or migration.
Environment variables

You can specify environment variables for your application. They can be edited after creating a service.

Creating Python service. Advanced settings. Persistent data volumes

Click “Show advanced settings”.

The first block in this area is Persistent data volumes.

D2C separates the application itself from its data. Docker volumes are used to store persistent data. Data is stored locally on the hosts. Any data which is generated by an application should be added to Persistent data volumes. All modifications outside of these directories will be deleted after restart/rebuild/redeploy of a container/service (Docker restores the original state of a container). 

When you have several containers and want to synchronise data between them you should mark “Sync” checkbox. We use Lsync for synchronisation.

However, we recommend storing user generated content in a cloud storage like Amazon S3 or CDN.

Creating Python service. Advanced settings. Configs

You can add your additional config files. These files do not change after restart/rebuild/redeploy of your applications. 

Click “+Add custom config” and specify a full path where it should be stored.

After that, fill it in.

Generating new config button, in this case, erase all content.

Creating Python service. Select hosts

You can choose one or several hosts where the similar containers will be deployed. It’s not necessary at the start, and you can scale your services after deployment. Also, at this step you can create other hosts and choose them even they are not online yet. The containers will be deployed when they are ready. 

After succesful deployment your project should look like:

What next?

In the next article we talk about:

  • creating an edge service for application services and add your custom domain with free Let’s Encrypt TLS certificates
  • scaling, balancing, updating sources, checking containers logs and metrics