98-365 – Microsoft MTA Windows Server 2008 – Server Performance Management

1. What to expect in this section

Server performance management In this section, we are going to talk about tools and applications that you can use if you want to monitor your server. It’s really important because you want to be proactive. When something goes wrong, you have to know where to start.

2. Hardware or software issues?

It’s time to check our server. The first question we have to answer and understand is: is it a software or hardware issue? If you see a report or receive an email, that is a question you’ll have to answer. We have already discussed storage. However, it’s not enough.

What you have to understand is what’s inside a server. It’s not enough to know that there is a SATA drive. Oh yeah, I know there is a Scazi connector at the back. It’s not enough. I appreciate that it’s not easy to start. That’s why I recommend them to my students. They go for a PC. It is very similar to a server. You will find a system board CPU, RAM, USB ports at the back, and things like that. and you can learn a lot. Of course, what you can do is much cheaper. If you want to work as an infrastructure engineer, I always recommend that you go and buy an old PC. It doesn’t have to be socket 1155. That is pretty expensive. It can be a pretty old system. I bought a PC for around £50. That’s absolutely fine.

The idea is to open it, see what’s inside, and say, “Okay, yeah, I can see this slot.” What is that? I can see this thing. What is that? Okay. Can I replace a CPU? What if I wanted to add some RAM? Of course, we discussed an HP micro server. That would be a great choice as well. You will find all these things, of course, on the system board of a micro server. Another thing you can do is just go and download some documents from a system board. Most of the time, you’ll find a really nice picture diagram with all labels. Everything that you need to know It’s great and very easy to learn. Let’s check. That’s what I found online. Yeah. Please note that everything is labeled. You can see. Okay, so I can see a CPU fan connector. Okay. So it means I have to route a cable and connect a CPU fan. Okay. Okay, here is my main CPU power. Okay, so here is a connector for my power supply unit. Okay, I have to route a cable here. What about Ram? Okay, I can see four slots. It is very easy to identify. You have all these sections described in more detail later on, including some information about whether it’s okay to note there in blue that that’s what you should probably instal here in Purse. That’s what is described in a document like that.

And it’s very easy to follow and see what’s inside. It will be very similar for a server. I think that is a system board from a server. I don’t really remember. It doesn’t matter. At this stage of your career, you have to have a very good idea. I assume that’s the case because you have two Nick guards. That is not something you would find on the system board of a PC. That is a service solution. A great way to learn You can take advantage of a document like that, and without opening a case, you can see what’s inside. Now, when you know hardware, software, and everything, In the next video, we are going to talk about tools and applications that you can use to check if your server is absolutely fine. Examine the performance. Make sure that your hard drives are okay. The CPU server has not crashed in the last couple of days. Thank you very much.

3. Understanding Performance Monitoring

Tools that you need to know We’re going to talk about the two main tools that you need to know to troubleshoot and check the performance of your server: Task Manager and Windows. Reliability and performance monitor Okay, let’s see. Reliability and performance monitor I am connected to one of my Windows Seven 2008 boxes. And here we will start with the reliability monitor. Let’s look into what happened. Here I can see that this application has stopped working. You scroll, and you can see. Okay, something happened a few months ago. Okay, here’s something that happened a year ago. and so on. It’s really nice. It produces a nice report. You should take advantage of that, check it from time to time, and make sure that you’re okay. And yeah, I am pretty happy when I scroll. I do not see 500 issues a day. That’s always a good sign. Now, this guy Performance Monitor is a pretty advanced tool that allows you to add additional counters and monitor almost anything you want on that server. I’ll show you what you can achieve. For instance, on your hard drive.

Okay, so let’s check the hard drive. And in here, let’s say I want to see bytes, write, and read. Let’s say I want this one. Okay? You can add any counters you want. Then it is on the list. You can disable the ones that you don’t want. It will start producing a graph that you can clear. You can save it and see what’s going on on your server. Again, you can monitor the CPU, memory, hard drives, nick cards, network cards, and almost everything else on your server. If you think, “Oh, I’m not happy,” just start with one at a time, and you will see. Be very careful with the scale because it’s not the same. That’s why it might be confusing. For instance, here, it makes sense to look at these two guys because they have the same scale. Play with it and see that in action. We right-click on the task bar, and we can access our task manager from here. Let’s start with services. It’s a list of services that are running at the moment.

What you can do, for example, is stop or go to a process, and from there you can end or set priority. If there is an application that you want to make really important to you, you can change the priority. We’ll look at applications that are running performance graphs on Windows Seven 2012. You will see that they tweaked and changed a lot of things. In “Networking,” you will see network utilisation and users that are connected to the server at the moment.

If you see a dead session, you can disconnect or even log off. Be extremely cautious when logging off, as he may have opened something that he desired. Windows Server 2012 Please note that it looks like Windows 8. Yes, here it’s much easier. I like the way they redesigned the taskbar manager. It’s much easier and faster to spot problems. For instance, here you can sort, and here everything will be highlighted for you. It’s really nice that you can see. Oh, yeah. 33%. Can you see that? Yeah, it’s really nice. I really like it. Then open the same tab. They redesigned the graphs so that they all looked the same. You can decide what you want to do. You can disconnect, then go into details about how to redesign everything on one screen. And of course, services where you can right-click, start, stop, or even search online.

4. Logs and Alerts – when things go wrong…

logs and alerts. Yes, you have to go through thousands of messages every day. If you work as an infrastructure engineer, you will see a lot of messages and have a lot of things to check. The main tool that you can use in Windows is, of course, the Event Viewer. Have you seen it before? I’m pretty sure you have. If not, let’s check it out. We go to “Start Administrative Tools,” and here you will find our Event View. It is a tool designed to collect all logs and important information. And I’m not only talking about things that go wrong; it can also collect things that are okay. And thanks to that, you can go back and see what has been happening.

I like to show that on a new server because you will not see 500 messages per hour here. Please note that we can see just five errors. when you click. You can go there and see what has been happening. Here we can see a DNS error, but even though it’s an error, that doesn’t mean it’s something really serious. When you add a new feature, for example, it can occasionally generate an error. and that’s fine. You can check it out, see it, Google it, and make sure it’s okay. If it’s just a one-time thing, you can ignore it. I recall one client who was using a very old version of Windows 7, 2003.

I remember that Windows managed to report an issue with the hard drive. And it was a good thing that we checked all the logs from time to time because we were able to spot it and sort it out. The hard drive was broken; we replaced it, and the client was very happy because we were proactive. In the real world, you will see thousands of lines per day, and it is going to be your job to learn how to use this option, okay? You have to play with it and make sure that you can recognise what’s really important. It can take some time. You will learn it; don’t worry. In our next video, we are going to stay in the same area for logs and alerts. We are going to talk about SNMP and some applications that you can use that are, in a way, proactive, can produce nice graphs, and email you when something goes wrong.

5. Monitor your server using SNMP

Let’s talk about the SNMP simple network management protocol. It will be on your MTA exam. They can ask you what that is and how it works. We will try to cover a little more than Microsoft decided to include in MTA. I will give you an example of an application that I have been using for many, many years. You can try it at home. In my opinion, the easiest way to learn is to try it out. Put it on your PC or server and play with it. That’s how I learned everything about SNMP. I started monitoring my network devices and servers. Believe me, in five days I learned everything I had to, because I wanted to include all possible features.

And then you have to understand very well how all these pieces work together. SNMP is a protocol for network management. Here we are going to use COVIDSNMP as a monitoring solution. There are two approaches. Two solutions. Let’s say this is your server, and then you have a box that is your monitoring server. Okay? So this guy is monitoring your network, and let’s say you have two servers. First of all, before we go and answer the question, there are two solutions. What do we mean by network management—monitoring all possible things? For instance, what you can monitor is the CPU, okay? You want to make sure that your CPU is okay. And if it goes above 50%, you want to receive an email. What about your hard drives? Make sure that you have enough space. If it’s below 10GB, send an email. What else? Your network card You don’t want to have a lot of traffic on your network. So let’s say if it goes above 15% utilization, then send an email. a very easy probe that you can create, of course. Is this server alive?

Okay, we can—what else can we do? We can, indeed, use the ping protocol. We use ICMP to make it happen. Okay, so a few examples will make more sense. You will see what we can monitor in a few moments on my server as well, because I have all these probes in place. Doesn’t the name imply a probe? Okay, going back to our example, we have two servers and one monitoring server. Now, there are two approaches—two ways you can make it happen. The first thing you can do is ask these servers to report back to that box. What I mean by that is that you can set it up so that this server will check its own CPU. If it exceeds 50%, it will reach this seven and declare, “Made alert.” My CPU is at 52%. What do I do? And this guy says, “Well, just leave it with me.” I will email our engineer and his MTA certificate. He will know what to do. The same applies to a hard drive, whatever you want. When something happens, the service says, “Okay, I just received this or that, or something like that, and you asked me to tell you about it.” So that’s the first approach. The second one is that you can be proactive. What I mean by that is that you can go and check.

So this monitoring server goes to the server and says, “Dear server, can you be so nice and tell me everything you know about your CPU?” And the server says, “sure.” Okay. So my CPU currently has four cores. Core number one is 5%. We are not really busy. Core number two is 7%, and so on. Then we go to that server, and we ask Dear Server 2, “Can you report your hard disc space, please?” And this service is sure; I have 75 GB available. Okay. And we do it, say, every 60 seconds. Thanks to that, we can create a nice graph, which we can see okay, yeah.

So it has been changing, going up, then going down, going up again, and so on. Yeah, we can produce a nice graph. Thanks to that, it can help us make some decisions. For instance, if you see that your CPU is like that, maybe it is time to add another CPU. A lot of services allow you to do that. You can add one more CPU. So, if you see that, let’s say you got GPU, it’s like, “Here’s 50%, here’s 100%, and this is when it all starts.”