How to monitor your Python Flask web application using Azure Application Insights

I recently worked with a customer on hosting Python Flask web app and web jobs on Azure App Service. When it comes to monitoring, Azure Application Insights is really awesome because it allows to collect traces, requests and exceptions very easily and build analytics queries and dashboard for visualization out of the box.

Azure Application Insights comes with a Python SDK that supports direct integration with the Flask Framework. To help you to get started quickly, I have made this quick start sample available on GitHub.

The most important parts of this sample are described below.

Thanks to my colleague Clemens for the precious help.

read more

How to use the Azure Container Instance connector for Kubernetes with Azure Container Service (AKS)

Azure Container Service (AKS) is a new service (currently in preview) that allows to deploy a managed Kubernetes cluster into Azure. Basically, you only have to pay for the nodes (virtual machines) that run in your cluster and you do not have to deal with Kubernetes masters. Actually, you do not even see Kubernetes masters that are totally managed by the AKS service.

Azure Container Instance (ACI) is a serverless service that allows to spin up both Linux and Windows Containers, without having to deal with complex infrastructure or orchestration system. Machines that run your containers are not visible and you do not pay for them. You only pay for your containers, on a per-minute billing base.

In this blog post I will explain how it is possible to use the ACI-Connector for Kubernetes, that allows to ask Kubernetes to schedule workloads into Azure Container Instance.

read more

Kubernetes on Azure: deliver applications continuously with Jenkins and Helm

Continuous deployment is essential in every software development project as it allows the applications being developed to reach the users as fast as possible, with the best quality as possible.

By automating all the steps that allow your applications to be automatically built, packaged, tested and deployed into several environments, you will drastically reduce the risk of having bugs reaching the end users.

In this article, I will explain how you can use tools like Jenkins and Helm to set up a continuous delivery pipeline of a containerized application that runs on Kubernetes.

read more