Azure monitoring setup

[RebeccaIlene]

Hi Support,

We are seeking a solution to monitor our Azure workloads using our existing Nagios monitoring system. We have our monitoring server in our DC with two Nagios monitoring VM in the Azure tenant, express route in place and Federated AD connect sync.

The cloud tenancy consist of IaaS based VMs; AKS, Azure SQL, Datafactory, function apps, log apps, app services, storage account etc in PaaS and also an O365 integrated SaaS service. We are looking for a solution to integrate Nagios to monitor below:

• Resource creation (especially blob storage with public access)
• Network changes
• High resource utilisation in IaaS and PaaS
• Service health monitoring in O365
• Critical alert notifications

Please let us know how can Nagios integrate with Azure? Is REST based API or Azure API available or do we need to use plugins? Are there ready-made packages we can use?

[jbrunkow]

I believe all of that functionality is already in XI once you have deployed it. You just need to navigate to the Configure section using the bar on top > click Configuration Wizards under Quick Tools on the left pane > and click on the Wizard relevant to the host and/or service you want to monitor. There is even one for Azure!

You could monitor the VMs using a variety of agents, but NCPA is the newest and available across platforms.

Let me know if you have any more questions about how to set checks up for any of your Azure services!

[RebeccaIlene]

Thanks for that information. This is for monitoring the VM.
However, we also want to monitor PaaS, SaaS, O365 and network changes.
Can you please suggest on the same?

[jbrunkow]

These terms are somewhat related, and tend to vary based on what cloud host we are talking about. Since we are talking about Azure, I will borrow some language from Microsoft to make the discussion more clear, and address each component.
*Microsoft on PaaS vs SaaS

• Hosted applications/apps

Any time you have a link to a web application, you can probably run the Website Wizard against it.
MONITORING WEBSITES

• Development tools, database management, business analytics

Again, these may be websites... We have several Wizards available for various types of databases.
MONITORING SQL

• Operating systems

Any operating system you want to monitor on the cloud is probably in want of an agent. NCPA works on Windows, Linux, & Mac. NRPE only works on Linux.

• Servers and storage

If it's a server, what's the operating system? If it's storage, can we run a database Wizard against it?

• Networking firewalls/security

Wizards exist for Network Switch / Router and Generic Network Device.

Data center physical plant/building[/list]

I would leave this part up to the people you're paying... After all, isn't that the point of SaaS?

You may also want to have a look at the Azure monitor on Nagios Exchange. It looks like that plugin offers Python scripts for monitoring Active Directory, Compute, PaaS, SQL databases, and Azure storage.

I hope that helps! Let me know if I can help configure any specific checks for things you depend on in Azure.

[RebeccaIlene]

I guess one option we have is to create scripts that can give us the required output and call the script using NCPA.

Will that work?

[jbrunkow]

Yes, NCPA can call a script using the -M option. Generally, additions to Nagios like this are called plugins.

Please refer to the following knowledge base article for more detailed instructions on how to call scripts using NCPA.
SCRIPTING PLUGINS FOR NCPA (KB)