I'll step first into the profile. So here's where, in this system, I've loaded in a couple of the profiles from the library. So we were looking before at that Windows profile. So if we want to use an agent approach, here's a profile that I can point and click assign to hosts. As I've done here, I'll go into assign hosts, and here you'll see the three Windows machines that are being monitored are assigned this profile. This is how I can rapidly get monitoring coverage to lots of devices. Right. So essentially, like you're saying, you have these hosts or servers or boxes or whatever out there, and you're saying, here's the grouping of parameters or attributes or whatever I want to monitor, the profiles. And you want to start monitoring that stuff on those three boxes that you have. Correct. I can apply this profile right from here and then activate the configuration, and I'll be processing those checks against those hosts. So Profile Importer is a library of where all of the possible profiles I may want to use are available. Well, I'm guessing from looking at the XML extension here that while there seems to be a fair amount of these that sort of ship out of the box, that people could import these during runtime or even make up their own profiles to import. That's correct. You can create, again, follow the open schema that the files are using, and then you can import them. So we saw before the Groundwork Connect site, if I had just downloaded an XML file that had some profile information, I could do that right here. So as an example, I actually just uploaded a couple of these. These are some new plug-ins that we're working with. I can upload them right from this interface and then load them. You know what I'll do? I'll redo this one. We should get an error. Okay, so I've already uploaded this, and it's telling you that I can't upload that.
2:07All right. So it doesn't let you overwrite existing things. Yeah, it wouldn't let me overwrite the one that was there. But this is actually the profile that was loaded in. And what this is is an SLA model that we've provided customers that allows you to group things together for service level monitoring. We get a lot of requests for cluster monitoring. You know, I have three or four MySQL servers or MS SQL servers. Don't tell me that one of them is down. Tell me that, you know, one of them is left. And that's something possible with these group checks that you see. Flexibility that people really like about this product. So whether it's SSH, whether it's SNMP, do I want to put an agent on there, or do I want to, without an agent, say, on a Windows machine, monitor an MS SQL? So you'll see that if somebody, it's one thing to say I want to monitor my MS SQL. It's another thing to kind of separate what I can do with open source versus what I should do. And here's an example where all predefined for you are things like watching for buffer cache hits, deadlocks, other waits, and so forth. That came from our work with customers in monitoring that particular application in production. Right. And is that typical that with the sort of, I don't know what you would call them, sort of default or out-of-the-box profiles that you have, that you might add different parameters for each version, sort of updating and refining them? Yep. So this is the grouping. So let's say, for example, we wanted to go into this buffer cache hit service. Again, before we would want to run our scan, we may want to be, we want to know what profiles we might automatically assign. So here I come in, and this is going to take me into a quick search through all of the services that are configured. So I'm going to drill down into the actual config. As you get familiar with the capabilities of Nagios in this screen, you'll see things like mouse over help. That's always useful to know what the different metrics can do.
4:09But now when I come down to the actual service check, here you'll see the values that are assigned in it. And in this case, this particular check is going to be giving, it has predefined thresholds of 80 and 50. So in this case, we're going to warn when the cache hit rates on those buffers are at 80%. Generally, you want caches to be at 99%, right? That means that the cache is being effective. So in this case, the value of 80% would create a warning, and the threshold of 50 would represent a critical. If I knew my databases and said, well, I want to know about it at 90, that's a warning if it falls below 90. And the critical is if it falls below 85. By making that change here and saving it, not every machine that has that service check can be updated through the profile. So let's now go to the discovery. So we started with Groundwork providing and actually contributing back an extension for Nagios to provide a web-based configuration. Natively, Nagios is a text file configuration. We then extended it with 5.2 and added some discovery features to it. So sometimes users will need a more rapid way to build a configuration, and auto-discovery takes that in a new direction for us. So for starters, let me first go in and talk first about the definitions. So here you'll see a basic template as part of Professional that's combining TCP and SNMP. I'm going to click on Edit on this template. And you'll see that I have defined a couple of methods along with the discovery. Now this is set for interactive, so I can watch what the system is doing, or I can automate the process of adding hosts and even removing them if I wanted to. I'll first stop on what we're doing with Nmap. So when I run the scan, it's going to use these variables that we've set up on Nmap.
6:12And we're also going to be checking SNMP. If I have SNMP read strings that I need to enter, I can put in a whole list of read strings. And this will help drive what services are going to be assigned automatically to the SNMP devices. Let's go and just look again at ranges. So I'm just going to do a quick scan. Now I've got an example set up. And the application does tell me that, hey, I ran a discovery. You've got hosts to process. I'm just going to click a Start New Discovery for demo purposes. Again, little things about making open source enterprise, things like, hey, you know, your security guys might not like you to run an Nmap. So a little warning there for you. You're about to freak out the network people and the security people. So get ready. Exactly. And again, fortunately, you can't tune what Nmap options you're going to use. Perhaps with Groundwork support, you may want to tone it down. But it's kind of mild, not really setting off a lot of bells. So it's gone through its discovery stage. Again, it only used one IP address. But it found a new host called CentOS. So it's going to resolve the hosts when it finds IP addresses. And it found that my host is called CentOS. And it did find that it is a Linux machine. You'll see it attempted to do an SNMP scan, but there were no hosts to scan. So as a host is discovered, we see the SNMP running. If it is, we would then have done a walk on it. We found our one host. We want to click Next. And now it brings me into a review. What did we set up on this host? Well, we set up the basic discovery. It's going to use a basic profile here. And what it found on this host happens to be three open ports where it assigned these three profiles. If I wanted to add this as it is, I can select it and then click Add. But before we do that, let's go in and take a look at the actual services.
8:18And if I now wanted to add new services, I can do so here. So maybe I have some SSH checks that I wanted to run on it or load checks. I can add those at this time. It also assigned it to the host group Linux servers. And I can also now choose specific profiles. And let's go back and just take what we've been given in the interface. Process that. And now we've just added that new host. And then I would assume it's going to, we'll start being able to see it in the status panel or whatnot. So you'll see it just came back. It reported success on the commit in the foundation synchronization. It reloaded Nagius. And now this host is set up for monitoring of those ports. Now we can get very fancy with what happens when it's finding a host there. And I'm going to come to the automation scheme. And this is where those decisions were made to assign the SSH alive and other items. All right. So basically what Groundwork is doing with NMAP and the SNMP walk is we're creating a text file. And then from that text file, we're learning out a schema that says in that text file, column 1 is the name. Column 2 is the alias. And then we can set up some task names. So as you'll see here on task 2, I'm going to click on assign name. And based on column 2, whatever that value is, as is, we're going to assign the value there to the object name in the configuration. In step 7, or I should say in column 7, is where I can specify that there's a service profile to be specified.
10:18Now we'll see that when it gets on that, here are the tasks that we can run against that piece of data. Right. So essentially, like you're saying, you have this text file with various columns. And now you're mapping it to the different aspects or attributes of that host that you're interested in. And then even, like you're saying, attaching new actions to it or tasks. Correct. Well, let's go back and check on our host that we just created. So the name of that host was Centos. And here he is now happily being monitored by Nagios. We've got two warnings already. And now we can see activity that's coming back. This is an example of groundwork correlating events for hosts together with the Nagios state. When I go into the event console, this is another new feature of version 5.2. This is where we have more of, again, NOC style type of an interface for all kinds of events. Right. The more traditional show me the river of events sort of things with, of course, filtering and all of that stuff. Exactly. Instead of it being only seeing events relative to the service or the host that we were looking at in the previous few. That's right. That's right. So here I can search and look for messages. And I can see, okay, here was a trap that came in with warm in the message type. Uh-huh. I can change this to a different operational status. And you'll see also the same host groups apply because I can now look, say, at my New York routers again from this particular console. If I wanted to look at all of the accepted traps, I can sort and sift them as I like. But more and more, customers are liking the approach of, well, I'm only going to be in the screen if there's a problem. And if Nagios can help me stay out of that screen unless there's a real problem, I don't have to try to digest lots of messages.
12:23Right. So what you're saying is people don't really like to be in the, and looking at events, right, it's nicer just to look at services and hosts, to get that higher level view of things. that's right