Microsoft Azure AZ-801 — Section 9: Manage failover clustering Part 2

61. Upgrade a node to Windows Server 2022

Now when it comes to upgrading nodes to a newer version of server, like, for example, let’s say that NYC-SVR1 was server 2019 or something and you wanted to upgrade it to 2022. There’s not really anything magic about this. It’s actually pretty straightforward. You can’t have the node be part of the cluster while all this process is going on. So, what you end up having to do is to evict the node, perform the upgrade and then you can reattach it. And so, essentially, what would happen in this case, since I’ve only got two failover servers, two nodes, all traffic would go here, would be connected to DC1 and then I would drain traffic from NYC-SVR1 and then evict it.

So, what I would do here is I would click here and go to Pause, Drain Roles, right? And if you then drain roles at that point, all traffic is now going to be hitting NYC-DC1. And then you would go here to More actions and you could stop the service if you wanted to and then evict it and evicting this removes it out of the cluster entirely. And at that point you could perform the upgrade on this server. Once the upgrade has been performed on the server, you can open up failover clustering and you can just re-add the node. So, you just go here and you can re-add the node. At that point you’re good to go.

So, again, there’s nothing really magical about this. It’s actually a pretty straightforward process. I am going to resume this, but there’s nothing, nothing magical about it. It’s just a matter of draining the roles over to the clients or over to the other server, and that just gives it a smooth transition. I’ll be honest with you, if you just try to evict it, it’ll do that. It you could see a slight delay with your clients draining over, though it may take a little a few more seconds for draining to occur. But if you drain it ahead of time before you evict it, then at that point your traffic’s going to flow over to that other server.

So, the main thing here to remember is draining. You would go through this process of going to pause and you would drain the roles first and then all traffic and all that would start hitting NYC-DC1 and then you would go through the process of you can just evict it, you can stop the service and evict it. That’s going to be the process for removing it. And then at that point you do the upgrade and then you reattach it and there you go.

So, again, nothing really magical about it. Pretty straightforward process. You’ve got to take it out of the cluster first, do the upgrade and add it back.

62. Manage failover clusters using Windows Admin Center

All right, so now I am connected into NYC-SVR1 and I’m ready to install the Windows Administrative Center. Now, if you’ve already got it installed, maybe, you kept some virtual machines from some other course or something. That’s fine, but I don’t have it installed in NYC-SVR1. The other thing to keep in mind, NYC-SVR1 cannot be a domain controller when you install the Windows Administrative Center.

So, how do we install Windows Administrative Center? We’re going to open up our web browser and going to type Download. Here we are into Google, by the way, Windows Admin Center. You could do this from being as well. Here it is, Windows Admins. And we click on that. And then we’re going to click this MSI download.

Now, I went ahead and did this just to save time because my connection is a bit slow, but here it is right here, the Windows Admin Center. And go ahead and click Open on that file. All right. And at that point, it’s going to pop up this little installer. We’re going to pretty much just take the defaults on this. There’s nothing really that we’re going to need to change a whole lot on this except the license agreement. Just the required diagnostics don’t need updates right now. The defaults. I’m going to leave this as default. All right. I’m not going to do the automatic update right now. I’m going to click Next. Default port 443. It’s going to use a self-sign certificate. That’s all fine. And I’m going to go ahead and click Install. All right. So, that only will take just a couple of minutes. I’ll pause recording while that’s happening. All right. Once that’s done, we’re just going to click this little link here. And this would be the link you would use to get to the wax server.

So, we’re just going to go ahead and click on that. It should pop up in our web browser here. And it is a little bit sluggish because I don’t have a ton of memory here, but I’m just going to go ahead and put my credentials in here. All right. We’re now logging on. And right now NYC-SVR1 is what we call a gateway server. So, it’s going to be our main server that’s kind of controlling everything. But we can add additional servers to our whack environment here, including we can even link to the cloud. All right. So, then it just pops up this little What’s New message. I’m just going to close out of that. Okay.

The next thing I want to do is I just want to add my DC, so I’m going to click Add and we can add the DC-server right here. And we’ll just put in NYC-DC1. And it will detect it. It might ask us to put our credentials in. So, we’ll go down here, put in my credentials, which I’ve actually already got saved. So, I’m just going to put that in and click Add. And then from there we should be able to click on it NYC-DC1 and take a look at it. And so as far as clustering goes, just be advised that if you don’t have a lot of memory. You can get some errors right here. It can throw some errors up saying it’s got issues connecting and all of that. So, just kind of a forewarning that wrack is another service that’s just taking up memory. We’ve already got so much going on these servers, you know, that it may struggle memory wise here, but it looks like it did connect.

The only other thing I’ll mention as far as that goes is, is if you did not have the clustering services installed, you could scroll down here and you can go to roles and features. Now, of course, with clustering and everything already installed, we aren’t going to really be able to set up another cluster, but it’s really the same process. There isn’t really much else for you to learn here, but I will show you where you would go to add a create a new cluster if you were going to create a cluster. So, you just go over here to a Windows Admin Center. And then you would click Add. Scroll down and click to Create a new cluster. All right. And it would go right here. So, select the workload Windows Server cluster where roles we aren’t doing virtual machines, would be clustered. This would be like an app service. In this case, it had virtual machines. You could choose that. And then all servers are in one site. If you do in two different sites, you would go that route. But I would click Create and then it says, “All right, make sure you got the following things checked here.” At least two servers with supported hardware, one or more network adapters with static IP per server management and separate adapters. And I know I have in my case, I’ve got I’m using DHCP, I’m not using static adapters just because of the way that we’ve configured our little virtual lab, but you could change it to static adapters if you wanted to. And then the Windows Server 2016 or later account, that’s member of the local admins group on each server. An Active Directory domain, an account that can join the servers. Configure your network switches for the VLANs required as Data Bridging. That’s if you’re going to take care of the memory requirements and VLANs. But we don’t really need all that.

So, then we would just click Next and then we’d have our credentials. We’d put our server name in here like nyc-svr1, and it would locate that, eventually it would locate that and then we can click Add. And at that point you could add server, you could add DC1 as well. Keep in mind, it’s going to throw an error because, you know, you’ve got cluster services and all that stuff set up.

So, but I’m just saying if you were actually setting this up for real and you hadn’t already gone the route of failover clustering, then you could go that route. And so there it is, right there, so that I could click Add and it’s going to just validate it in a way I go. But when I try to add the DC, it’s going to fail just because we’ve got services and all that stuff going. But that is the process. It really isn’t very different. You’re just managing this to a web interface as opposed to failover cluster manager.