Azure API Management (APIM) service is one of the building blocks for AI applications. At the same time MCP (Model Context Protocol) server is one of the latest capabilities around AI. Not so long ago APIM announced support for exposing either existing MCP servers or existing APIs as MCP servers. I have played around with this functionality lately and what it turned out is that these MCPs Servers in APIM are just API resource underneath which means it can be easily deployed with Azure Bicep. In this blog post I will show you how to do it with examples.
Tag: Cloud
Monitoring canceled Azure Subscriptions
Azure Subscription cannot be just deleted. They go trough different states. Although you might have taken all necessary to secure your Azure tenant sometimes mistakes happen or your environment might be compromised by bad actors. The first step of removing Azure Subscription is to cancel it. That is critical action that you may want to monitor although you should have in place other ways to monitor resources that will signal that they are not available. In any case additional alert that this action was done could be useful information to be alerted upon. In this blog posts we will take a look how we can do that by using Azure Monitor. The deployment of the Azure Monitor resource will be done via Azure Bicep.
Azure Bicep fail() function explained
Azure Bicep fail() function was introduced in Bicep CLI v0.33.93. In short this function allows you to fail your deployment when certain conditions are met. In this blog post we will look at how to use it in real world example.
Using Optional parameter if not configured in Azure Monitor workbooks with KQL query
Azure Monitor workbooks are great way to visualize Azure data for monitoring and analysis. Although there is a good documentation on how to built them I would say that examples for more advanced scenarios are lacking in there. Recently I was asked a question about such scenario: “How do I set KQL query in a way that the filter for optional parameter in Azure Monitor workbook is not applied if value is not provided for that parameter?”.
Testing Data Processing Azure Bicep Functions Easily
These days Azure Bicep has a lot of more functions that can be used for processing data. Most notably I am referring to the lambda Azure Bicep functions. Often times I use two or more of these functions all together in Bicep templates. When using several of these functions chained one after another it is easier to get lost how data will be processed. Also to test all the different input that will be passed to those functions will results in doing several different deployments. That is time consuming process as deployments takes time to run not to mention that along I have several other resources also deployed via those templates. Thankfully just to test if the data is processed according to how I have imagine it there is easier method by using Bicep parameters files.




