Each micro service has its own database in order to be decoupled from other services. This the Database per Service pattern. Related patterns discussed are the Saga pattern, event sourcing and the transactional outbox.
In business-to-business scenarios, mutual TLS (or mTLS) is highly useful. It ensures traffic is secure and trusted in both directions between a client and server. The organization implementing mTLS acts as its own certificate authority. mTLS is often used in a Zero Trust security framework.
A Single Page Application is a literally an application with a single static page. In Angular the dynamic sections of your UI are referred to as microfrontends. Each microfrontend is an Angular CLI project that can use its own version of Angular and be deployed separately from any other.
NGINX is open source software for web serving, reverse proxying, caching, load balancing, media streaming, and more. It started out as a web server designed for maximum performance and stability.
The Azure Web PubSub Service can be used for real-time messaging applications using web sockets and the publish-subscribe pattern. We use subprotocol json.webpubsub.azure.v1 to exchange messages in JSON format. The actual xml message is a passed as a string via the data node.
PaaS vNext means you can deploy Azure Web Apps, Functions, Logic Apps, API Management and EventGrid to an Arc-enabled Kubernetes cluster, so not as a PaaS offering in the public cloud. Arc-enabled means that you can run the cluster anywhere: in an Azure VM, in the Amazon cloud, on the Google cloud platform or on premises. All in your own private compute space.
There’s a new interesting feature to secure your Azure Function via a Bearer Authorization token (preview). Use the OpenAPI Extension for Functions. It’s interesting if you don’t wanna use API Management.
Ideally Integration and BI work together on data ingestion, especially when the primary applications are outsourced to independent software vendors. But most often there’s no such cooperation. How come? Looking at the Lambda and the Kappa Architecture can be an interesting angle to reflect on the issue.
Architectural work often isn’t assigned to sprints. It’s better to put epics, features and user stories on the backlog and leave it there. No rocket science, but it saves you a lot of work.
Example of how to validate a Json Web Token (JWT) and retrieve a value from that JWT via an API Management policy.