More and more services are moving to the cloud, Azure is one of the larges players but AWS and Google Cloud are also two large players. But just because resources are moving to the cloud doesn’t change the fact that we need to know how our environment is doing. Since I´m a monitoring guy, I write a lot about Azure Monitor and the capabilities of it to help us monitor our resources in the best possible way. But there is another aspect I want to touch as well, Azure service health.
While we monitor our resources using Azure Monitor, who monitors Azure Monitor as a service? Microsoft of course monitors all the Azure services to keep track of the status and to take immediate action when something goes down. We have the possibility to check up on Azure services from within Azure Monitor and that´s what I will be telling you more about with this post.
The available Azure Monitor data sources is an interesting topic. Azure Monitor is a really powerful monitoring solution solely based in Azure, with a lot of capabilities. When the now retired Operations Management Suite were first presented, it was presented as a cloud agnostic solution meaning you could place your resources in any cloud besides from Azure, such as Amazon AWS or Google Cloud and still being able to monitor the resources. This is of course a real good thing (and necessary) since not everyone has or will have all their applications or servers in Azure only, there´s still a lot of on-prem servers and applications but also a lot of resources deployed in other public clouds as well.
Azure Monitor is a quite new addition to the monitoring sphere when talking about monitoring Microsoft technologies. Traditionally it has been System Center Operations Manager (SCOM) that´s been the go-to guy but with the new addition of Azure Monitor some things have changed. From time to tome one can hear the phrase “SCOM is dead” and that you should go all-in with Azure Monitor instead. But is it really that easy?
In my opinion, no it’s not. While Azure Monitor has a lot of strengths being cloud-based with regularly updates and additions, it still lacks some things that we´re used to from using SCOM for all these years.
A while ago we were involved in a project with one of our customers where our goal was to connect a large amount of servers to Azure Log Analytics. They had already done this with a connection through SCOM, but when they added another management group to their servers, so called multi-homing it stopped working. This was a huge issue since the data they were sending to Log Analytics were really important to them, and we started discussing how to do this the best way. We eventually decided to go with a direct connection to Log Analytics, instead of going through SCOM as they had done before.
on January 15th was the last day of the OMS portal before its
retirement. It has now completely moved to the Azure portal instead.
Since Operations Management Suite (OMS) have been retired for a few months and is no longer available for new customers, the portal had served its purpose and have now been retired. Nowadays administration of the included services is handled through the Azure portal instead.
A few weeks ago, we (Approved) held our annual event called SCOM Day in Gothenburg with about 80 attendees. This year we focused on hybrid monitoring using SCOM 2019/1901 and Azure. It was a full day of sessions where we had Thomas Maurer talking about Azure Stack, Martin Ehrnst who was talking about API integrations in SCOM. And lastly, we had Marcel Zehner who showed us how he monitors and interacts with his Tesla using Azure.
I also held a session along with my colleague Jonas Lenntun about the news in SCOM 2019 and 1901 where we focused on the parts that we think makes most sense and will most likely come to use for most users. The news was announced during Microsoft Ignite that took place in Orlando in September. But to those of you who didn’t attend any of these events, the news is still important to know about.
Lately I have been working a lot with monitoring VMware using SCOM for some of our largest customers and have gotten to think about this more and more. Even though cloud providers such as Amazon Web Services (AWS) and Microsoft Azure keeps on showing great numbers of growth (and profits for that matter), the absolute majority of customers IT are still on-prem. Since about ten years, virtualization has been about the coolest thing there is and the largest player in this area is still VMware.
Thinking about how large this area is and the importance for the organization, we need to monitor the VMware platform. Just as well as we need to keep track of what’s happening with our services, such as web shops or other business critical systems, we need to monitor the foundation it all relies on as well.
Back in January this year I wrote a post about how you can install the OMS agent using PowerShell. Now the time has come to include the Service Map agent in the equation as well since this is a feature that recently got Generally Available. You can find the original post about installing the OMS agent here. What´s new in this script is that I have added a section for downloading and installing the Service Map agent as well. Enough talking, let´s get to it!
One of the things I work with in my role as a product manager for Operations Management Suite (OMS) is the automation part of the suite. In this case, it means Azure Automation that can do a lot for us in terms of automating our recurring tasks. This post will be the first post about what you can do with Desired State Configuration (DSC) as a part of Azure Automation.
Before we get started there are some things worth knowing. As a part of OMS, the licensing for DSC is based on per-node and the listing price is at $10 per node/month. This means that each server you want to configure using DSC is assigned this license.
Before we get started there is one prerequisite you need to take care of; the latest version of WMF 5 (Windows Management Framework) needs to be installed on the server about to be configured as a DSC node. This makes is possible for the node to communicate with Azure Automation. You can find WMF 5 here. This isn´t necessary if you’re running Windows Server 2016 as I will be doing for this post.
The first thing we need to do is to create a file stating what to communicate with and what to do. This is called a MOF file and is what makes is possible to retrieve configuration, but also to register the server as a node to Azure Automation DSC.
A while back I wrote a post where I went through the basics and how to get started with the new feature Service Map, which is a part of OMS nowadays. Read the post here. That post only showed how to get started and what kind of information you will get from the solution, but it didn´t say anything about troubleshooting. And at last, it did just explain how to get started with your Windows servers (or clients for that matter).
Today I will go through how to get up and running with Linux servers as well, as well as some troubleshooting.
When considering the Service Map agent, the following Windows operating systems are supported.
- Windows Server 2016
- Windows Server 2012 R2
- Windows Server 2012
- Windows Server 2008 R2 SP1
- Windows 10
- Windows 8.1
- Windows 8
- Windows 7