Resource health through Azure Rest API

As a part of a large monitoring project involving on prem, Azure and Office 365 I have started to explore the different methods on how to acquire all relevent data. Previously, I have written a post on how you authenticate against Microsoft’s partner center API which is a part of the same feasibility project.
Later down the road i will try to write a larger blog series on how we can monitor Microsoft Cloud and on-Prem resources. Hopefully it will be joyful…

This blog post will describe how you can use Powershell to authenticate and get availability status from all resource groups and their resources. If you’re not that interested in monitoring data, use this post as a guide on how to get started with the API and the rest is documented on the Azure API documentation pages.

High level overview:

  • Set up an Azure Active Directory Application to authenticate (not covered)
  • Build an authentication header with a token from Azure AD
  • Get all resource groups within a subscription
  • Get the availability of all resources within a resource group

 

Get Azure AD application token

After setting up/registering the application in Azure AD you will have to use the application ID and secret in order to generate an authentication token to use against Azure management Rest API’s. I have created a basic Powershell function you can use, including an example authentication header.

Get all resource groups and resources

Next we will grab all resource groups within a subscription, before looping through each resource group to get the individual resources.

Now that we have all resource group and all resources within we can use this to further get the resource health and availability. To be honest we don’t need to get RG before checking the resources it self, but as a starting point i find it very useful to have the resource groups and their resources available within my scripts.

 

Get the resource availability status

Now we have authenticated, grabbed all resource groups and their resources and it’s time to find our resource health. There are multiple ways on how to get this data, and we have to consider the methods we use. One limitation i hit pretty fast working with this is the number of subscription resource get requests, specified in the response header “x-ms-ratelimit-remaining-subscription-resource-requests: ” All limitations documentet,  here 

When you hit this limit every substantial requests is dropped. I have to rethink my whole monitoring scenario due to these limitations.

Get Resource health based on resource group

This let’s us get health state on all resources scoped to a resource group. If we reuse the data from previously (we have all resource groups within our subscription) we will get all resource group health this way.

Other examples are to get availability by Resource and the entire subscription

 

 

 

Martin Ehrnst
Systems Engineer working with SCOM, OMS and Azure
IT Pro with a passion for monitoring. Working with System Center, OMS, Azure and related software and cloud services.

Direct customer experience from previously being a Technical Account Manager.

Community supporter where I try to contribute via blogging and social media.

Posted by Martin Ehrnst

Working as a systems engineer in one of Norway's leading enterprise cloud providers. Mainly working with System Center, Azure and Windows server products

*All post are personal

2 Comments

  1. Hey Martin! All the code samples are centred, what’s happening?

    I wonder how difficult it is to do the token requests without the Azure PowerShell commandlets 😉

Engage by commenting