When you are working with an architecture that includes Kubernetes clusters you might still find useful to use Cloud Run for certain applications, like a web frontend, API gateway and more. At the same time, you may find confusing how to integrate it seamlessly in your CI/CD workflow.

Modern developing usually includes a lot of steps that an application should pass before being deployed to a production. Such steps include creating an environment to compile an app, packing the app in a container, creating a testing environment and passing the tests, creating an environment to deploy to, doing a deployment and more.

When you are starting to have more and more steps, the complexity of steps is increasing, you are starting to use a microservice architecture with many apps, your dev team is growing. Then, to handle all of these you need to automate the process. This is CI/CD (Continuous Integration and Continuous Delivery) workflow is made for.

Designing a right architecture for your application can be a pretty complicated job. In this article I am going to get through our steps of choosing a platform and an architecture.

While deploying containerized React front-end for our applications, my business partner Iskandre and I have discovered that passing environment variables such as a listen address, a port, etc. without rebuilding the entire container is a bit not trivial, but possible.

Here on Iskandre website you can read the article how we did it

It’s a good practice to avoid storing any sensitive values in environment variables. So, my business partner Iskandre and I have integrated secrets usage in all our applications. The best way to do it in GCP is by using Secret Manager. Here you can read how to add an access to Secret Manager for your GKE deployment, so you can use secrets in a manifest.

Then you need to add an integration to your application.

Here on Iskandre website you can read a tutorial how to add Python Secret Manager to Alpine Linux containers

My business partner Iskandre and I had been choosing a fast webserver that we can use with Flask API. After reading some articles we decided to try Gunicorn WSGI + Meinheld. To learn how it will work on practice we did some tests.

Here on Iskandre website you can read the article