foreign so we're all here today to go over the cmdb health dashboard and if you're not familiar with what it is it's basically an easy way to evaluate your cmdb relationships and kind of your data everything involved in your cmdb kind of pipes down into some easy at a glance reports that you can go ahead and modify you can configure different ways so that you can get value out of your cmdb and actually see where things are kind of going wrong we're going to go over a handful of things right so what is the cmdb health dashboard from a high level what does it look like what elements are on it and then we're going to break it down into pretty much three different sections which is completeness compliance and correctness as well as relationship Health at the end after we kind of talk through a little bit of that stuff what it takes to keep the CND healthy and at the end questions and answers I will say too like there is no demo piece of this when you look at it in the agenda that's because I'm going to kind of take a break in between each section to show you what I'm talking about I think it makes sense to kind of talk about it look at it and go back and forth so that's what I'm going to do all right so let's start with asking what is the cmdb help dashboard kind of like I said right it Aggregates key performance metrics it offers actionable insights and it's a place to kind of Monitor and improve performance for cmdb that all sounds great right but how do you use it what value is it actually adding to your organization right so these key performance metrics and whatnot they look good at the start because you need to configure a bunch of stuff and how do you actually take what that's giving you and actualize that into you know an effort that somebody has to make to maintain things so why do you use the cmdb health dashboard so it's critical to success obviously as it says from an I.T cost perspective you know servicenow likes to shoot its own horn so it can save you about 40 percent roughly in it costs it's critical to avoiding unplanned outages data quality and the cmdb it's essential goal of any it organization I think we can all agree on that and it's kind of the black hole right you pipe information into your cmdb and a lot of people don't know what to do with it they think data is coming in that's good enough and what do you do with it once it's in there or how do you evaluate what's looking at what's actually in your system so you can proactively manage issues as well this is probably what I like the most about it you can create tasks out of a lot of these metrics that have people go actually remediate things like you know duplicate CIS or what happens when you're not getting required fields and you know what happens if you have orphans in your relationships things like that in your CI so it's pretty useful it can help notify people and make sure that you're improving your accuracy to your cmdb usability and act on any issues before they become major outages or anything like that and it offers an easy at a glance view for stakeholder transparency sea level people like to be able to see things in an easy way this gives you fancy numbers and an easy way to evaluate kind of the overall health it drives resolution for compliance failures and just basically tracks the overall health of your cmdb over time as well so those are all valuable things let's jump into the elements of the dashboard and and this is where I'll probably stop and kind of show you some stuff along the way here because you know who wants to really see a list view of things I put together all right so there I wouldn't say three different dashboards available but there's kind of different views of the cmdd dashboard and you start with the cmdb view there's a group view which you can configure different groups and filters essentially to look at things um I'll show you that in a second how you build that and what that looks like there's also a service view so you can kind of instead of looking at Everything At a Glance like that's what it does out of box right it kind of shows you here's all your Ci's and it shows you everything there's like a really basic way of filtering stuff but you can get more granular with it and that's basically the purpose of these dashboard views and there's also several different elements that make up the dashboard which is these tabs that I have laid out there on the slideshow for you which is cmdb Health relationship health and seem to be a health dashboard jobs so without further Ado let me pop up my Dev instance here so to start up here in the application navigator type in cmdb dashboard and you'll see these three options that I was telling you about right seem to be view service View and group view let's just click the top one cmdb view here and you'll see kind of what I was talking about all right so here are the three tabs that I was mentioning cmdb Health relationship Health you'll see each one kind of has its own all of its own metrics and whatnot I'll talk about this one later and then they seemed to be Health dashboard jobs so let me just take a second to talk about this so I believe out of box these are all active equals false so when I got my developer instance I went ahead and I turned all of these on these are your scheduled jobs that populate data in your actual dashboard so step one would be to turn these on I think the only one that was on was this top task generating Ci's calculation everything else was off so just go in here active equals true and most of them I want to say are scheduled to run uh let's just take a look daily zero three hundred hours okay so it runs pretty much once a day you can execute these at any point in time and as I walk through some of these things I'm going to show that to you so the first thing if you want to populate data right out of the gate make sure that they're all active come in here and execute each one of them from then on they'll run once a day which is you know we all know how scheduled jobs work anyway so let's go back to this cmdb health tab here and kind of talk through some of these elements that make up the dashboard there are they seem to be hierarchy here right when I was talking about some basic filtering notice right now it's all so it's all CI classes if you really wanted to get more specific just to say a Windows Server you can look in here and if this looks familiar it's just like your CI class manager here so let's just click here and then you can select a class and it will update all of this stuff based on what you choose here so you can dial it in pretty well I guess if you're specifically looking for a certain class that's all great however there are better ways of managing this let me take a step back here and make sure that I'm following along with my presentation all right so I talked through some of this stuff here let's go to the next slide and talk about what results are displayed all right so there's a cmdb hierarchy for filtering results which I just showed you there's the cmdb overall health scorecard which basically takes the combination of your completeness compliance and correctness score and rolls it all up into one nice looking number if you have everything configured that's great and it will actually add value if you don't it's going to look much higher than it actually is so if for any of us who have been asked to do a cmdb assessment you come in here and you look at it and everything's green and looks good for the most part that's because there's a lot of configuration that needs to happen behind the scenes to make this actually valuable all right so that was completeness compliance and correctness it'll show you things like duplicate CIS orphan CIS and stale CIS these are just going to be bar charts I'm going to dive more into this a little bit later as well because that has to do with a whole nother section called correctness and then there's top generators for incident alert and change and that sounds just like what I think what you think it might sound like um it's basically a list view of your Ci's your top offenders who creates the most incidents uh what's the eyes create the most incidents alerts and changes the alerts just a note that's for event management so out of the box you're going to see that that's blank until you turn that on and then you would have to have alerts configured and everything's set up there to make sure that that's actually pulling in data so there's several different ways that results are configured and created just a high level overview here I talked about the health dashboard scheduled jobs a little bit you'll see me go back in there I'll talk through that handful of times and you'll start to see how that all ties together you have your score generators and rules that 58 the top right of the screen that you're looking at that's basically a score generator and the rules around it you can say what the thresholds are so it's showing up orange and I think that out of box parameters I I don't know what they are up top my head I'll show those to you but it's uh you basically can set red orange and and green and yellow and kind of these different colors corresponding to you know the health of your cmdb and I don't know if people actually get in there and mess with that but I'll show you that as well and the CI class manager there's a tab in there called Health which you can see completeness compliance correctness and health inclusion rules so before I jump into completeness let's take a look at some of these things that I talked about right I mentioned the group View and then I want to talk about health inclusion rules let's go ahead and jump in here so you type in cmdb dashboard and as I said before there are these different views here so if I click group View if I don't have any groups configured you're going to see this little message that says no Health groups currently created go to the cndb groups to create one and you'll see all this stuff is blank here so just like it sounds I'm going to type in cmdb groups open this in a new tab and I'll come to this list view here I added this this column here group type now when you create a new group what you're looking at and I'll just name mine windows and Linux servers there's different group types here the only one that'll actually show up on there is if it's health so you can give it a description just for the sake of time I'm not going to but I'll submit this and you'll see it's the only one that shows up as health so if I don't Define a query here it's not going to actually show anything so click back into it you'll see that this area shows up here so I can create an encoded query and what you're going to do is just build out a couple of these right so I said windows and Linux servers and you can see here right now the only thing that's mandatory is your class so I'm going to set this Windows if I can type Windows Server submit this one and I can add as many of these as I want to so it's just going to basically aggregate everything together so that you can get a cohesive view of different groups of CIS all right out of those guys now if I go ahead and reload well if I use a shortcut refresh this and then it should show up here now when you create a new group like this it'll show but you'll have to rerun different jobs which you can see the cmdb help dashboard jobs here cool so you can see that this shows up now if I add a handful of these then you can go ahead and choose them so this is going to be useful for end users right so that they are interested in load balancers or they're interested in computers specifically they can create these different groups so that they can view the information they actually want to see when maybe just the basic filter didn't work out for them so if I want to populate all this data again going back to the the three there's the completeness compliance and correctness so if I come here to the help dashboard jobs you'll see their name's pretty easy to understand completeness compliance and correctness so I'm just going to open each one of these in new tabs and just kind of show you what that looks like you'll see me click around all I'm doing is going to each one of these guys and I'm executing the scheduled job so when I come back here once these jobs get completed see completeness already populated there a couple of these others kind of take a second by compliance and correctness take a second to actually populate but once they do see that and this is just for your specific windows and Linux and then once this one as well correctness almost always takes the longest because it has the most work to do once that's updated you can also refresh this guy too so you can see and that this will get value for people who are specifically trying to refine things down and again if these dashboard jobs are active they'll run once a day as well so that's how you create those groups all right let me talk about how to exclude some stuff really quick so I'm going to show you a health inclusion rule so we'll go to CI class manager this is kind of your bread and butter here where you're going to do a lot of the configuration if you're not familiar with CI class manager it's pretty powerful really I suggest looking into it if you're into the cmdb at all you're probably familiar with it if not it's a good way to kind of get information at a glance so I'm just going to again Windows Server because it's a good example and let's kind of take a look at some of this stuff so right here you have this tab health so if I expand this you'll see completeness compliance and correctness and I mentioned the health inclusion rules I'm not necessarily going to create one right now but basically what this is saying if I create a health inclusion Rule and I say I don't really care about recommended and required fields for Windows Server I can go ahead and save this guy and what this it's almost more or less a exclusion rule so if you if you create this filter and you add these selected Fields it's basically what's not showing up so there are times where maybe you're creating specific orphan rules where you're like you have a customer who does something kind of weird they're like hey we have a weird relationship maybe it's something like I don't want to see certain servers hosted in this specific data center because I just know that data's not there yet so you could literally go into whatever class that relates to and then kind of walk down you know and dot walk down to the right field and say I I don't want to see it and then you add this and you save it and then when you rerun all these scheduled jobs this isn't going to show up anymore on that dashboard so I've had a couple of clients to ask for that too completeness let's kind of get into the nuts and bolts here of how to actually use this thing and configure stuff so completeness from a high level you have two things that you need to be thinking about required fields we all know what required fields are they're mandatory if you go into the dictionary of you know any record essentially there's the mandatory flag this is really what this comes down to a couple of things to note about required Fields when Discovery runs it actually looks at this mandatory field and takes it pretty seriously so if you say hey I need serial number required it's mandatory and it doesn't have one it's not going to create a CI if it doesn't have that there is just let me add a note to that there is actually a system property where you can overrule that that's it's in their documentation you can find it pretty quickly but just just know that that's a thing so if you go willy-nilly and you say let's make all the fields required if uh discovery's not getting it then it's not going to create CIS so maybe don't start there so the next piece of that is recommended Fields this is pretty nifty because well nothing out of the box is specified but recommended is kind of like your testing ground for fields that could become required in the future so say that hey it would be great if serial number was mandatory we're getting a bunch of garbage data Maybe you go into that specific class and you know again Windows server is my kind of example I keep using but you say hey I want serial number on there I want to assign to on there uh maybe even support group and you can go ahead in the CI class manager you can add these you can see in this little pick list here you can move it over and those recommended Fields will then be evaluated on your scheduled job so let's again take a quick break from this slide and actually look at the practical application of this go back to our dashboard all right and everything that I'm talking about right now relates to this piece so cndb completing a scorecard 100 looks pretty good again remember that recommended isn't configured out of the box so that's going to show up as 100 right away if you have no required Fields that's also going to show up as 100 so if you do no configuration this is where this whole thing can be a little tricky it's going to make it look like your data is really good when it might not be so let's look at a couple of different things if you click this number right here 100 it's actually going to open it up into a more detailed completeness dashboard View so this is where we start to see information that's actually relevant CI is missing required attribute no data display no data to display over here and recommended that's because nothing's been configured so CI class manager again it's where we're going to go we've all made Fields mandatory so let me just uh hopefully you won't be too bored while we're looking at this but I think it's always good to see how things tie together let's go on Windows server and here we can look at a couple of different things so if we're looking at mandatory Fields right I click on attributes and you can see right here from a list view it'll show mandatory is false as we already know from those reports there are no mandatory Fields so if I was just looking here to say serial number all right it's this guy click the little I it's going to open up to the dictionary entry all right and you can see here the mandatory is false just note as well that this is specific to the class right so if I did this on the server level not Windows server but server it's going to inherit this field all the way down to all the child classes underneath of it Windows Server it will only be mandatory for here so click here to edit it and I'll go ahead and I'll make this mandatory I probably wouldn't suggest starting here by the way start with recommended Fields first so let's come back down here to the health tab click completeness and then you'll see a few things right so I mentioned these scorecards and rules previously and said who's ever going to configure this well if you were going to configure it this is where it's going to be required field scorecard once this loads you'll see that there's just basically a minimum and maximum threshold which corresponds to color coding and if somebody said you know I really want my required fields to be super efficient and maybe you increase your threshold to say hey if it's below 90 that's a red so you can color code things let me just show this requires field scorecard if I click new see if this is the guy I wanted to show you right here so if you wanted to adjust this from green orange red you know you can basically just come in here and set a value to say this is just saying hey if 67 percent of your your stuff isn't isn't compliant with this then it's going to be red you know we'll understand threshold so I won't beat this up too hard but this is how you would go ahead and adjust the color and at what percentage it changes colors on that dashboard so for anything that has a color on it and a percentage there is something like this in here right so it's going to be the same for compliance and correctness recommended field scorecard all right so I'll click on recommended Fields so this was like the screenshot that you saw on the other screen so if I wanted to just add a couple of things right that'll say assign to and then what's it owned by I'll just add those to save cool all right let's go back to my dashboard actually I can just run it from here so keep in mind what I said before too I'm going to come back here I'm going to keep running scheduled jobs so it's this completeness score calculation right required and recommended is what I've been touching so we're going to go ahead and rerun this job otherwise results won't show up until it runs at three o'clock in the mornings let's go ahead and run this guy so after that loads we'll start talking through compliance here in just a second because this is actually where stuff gets a little bit more useful I feel like all right here we go add some stuff around so CI is missing required attribute you see how that populated now and it actually has some good information you'll see that this all again has to do with win server the missing recommended because that's where I did that one in the dictionary entry I must have made a mistake there when I was doing that I must have been in server anyway okay so when you click into here it'll show your health results so it's pretty cool I'll just click an example of this it creates all these records in here so all I did was click into that bar chart and it opened up here so here are all the records that I have to do with that and I can see the following required fields are not populated for serial numbers so if I had a whole bunch in there it's just going to show me what each one of these are notice that there's this task field as well so if I wanted to I could go ahead and search for a task and relate it to that or you can always click new and you can assign it to somebody as well if you wanted to do that so that way people can actually take some kind of action with it which I find is pretty useful and it's going to be the same with recommended Fields as well you're gonna if I click into this the recommended attribute it's going to give me the same description of this as well which is just going to say hey this is missing it's going to give you a list and say following recommended fields are not populated so if you wanted to make something mandatory and you're running through Discovery and you're like hey you know we really should have this information let's make sure that we know what it is this is where you would probably start recommended Fields you to add them there and then you can just go each class at a time you probably wouldn't want to be too General with it you'd want to start with specific classes which recommended Fields there are and then you can actually get some useful data from there so moving on to compliance back to the cmdb view so that was completeness and the scorecard what makes that up compliance has to do with Audits and this is a pretty cool feature because the useful application of this think of things like if you wanted to find machines that weren't running anti-virus or a specific software right somebody you know an organization manages their software using a specific piece of software you want to make sure that's on all your computers you could create basically an audit that will go out there look at all your devices again based on which class you're looking at Etc and then it will pull back those audit results for you so what makes up compliance or a few things so the data basically the question it's asking is the data doing what you need the CIS to be doing right and there's four different pieces of this that you're you can really use to find this information which is a certification filter certification template and audit and scripted audits scripted audits will be used in the dashboard it's a good good use case if you have something that's a little bit too complicated where just a simple filter won't really get the job done scripted audits can be used otherwise if it's pretty basic just stick to the certification filter template and audits a couple of notes about this so this is one of the areas where you can actually see on the dashboard it is populated right 11 somebody might come in here and go wait a minute why is this 11 so there are actually a bunch of out of the box audits that are on and will be running so you should look at those and say maybe I don't want those on or you know maybe you do but either way you should evaluate those to make sure that you are actually getting the information that you want so I'll click into there let's go back to my okay all right so like I said deactivate any out of box audits until you review them and you'll have to create your own audits in here if you really want value from them you can also create tasks directly from an audit so say you had you were going to do an upgrade pushing out the newest version of you know a Windows Server operating system because that's been my example yeah and you want to make sure that you're auditing it to say I need a specific amount of ram I need X number of cores I need to make sure that this box is this old or whatever you can run an audit to go ahead and look at those values and make sure that you know all of your devices have as much RAM as you need they have as many cores as you need and if not it will show up in this audit list and you can assign that to somebody to say go ahead and remediate this or decommission the server and let's spin up a new one whatever it is you want to do this is where you would do that so I'm about halfway through on my time so I'm going to try to just show you some of these Audits and what they look like and kind of show you certification filters and stuff pretty quickly so I don't just lose all the rest of the time here to this one section but you can see Ci's with audit failures so let's I think what's helpful for this is actually kind of walk backwards look at the end result and then walk back and see how you got there in this example so let's take a look at some of these results that we're looking at all right so I'm just gonna again these are the out of the box guys I didn't really create any audits but you can see here this is basically what it'll do is it'll generate this health result and it'll tell you what class it'll tell you the description so failed not it for Windows computer minimum statistics tells you the date RAM and needs to be graded in a thousand discrepancy value is 503 so what we know is again you can come in here and you can assign a task if I go into this audit itself then I can actually automate this to create tasks I can assign it to a group I can assign it to individuals whatever you want again that's from the audit and this this is like super customizable or configurable you can choose any number of field so now that you've seen this let's kind of take a step back to see how you get there so again our old friend CI class manager is where you're going to probably do most of this I will say though this is one of the examples where if you type in compliance you'll see audits templates filters gives you an overview I think this kind of Blends different audits as well I'm not going to pretend I'm an expert on auditing if you wanted to actually look in there and see your different audits or templates from a list view this is also a really easy way to do it otherwise you're going to be in here in your CI class manager if you want to get specific so let's do Windows server open it up health and now we're on the compliance tab here so you can see I was talking about certification filters certification templates and audits all these different things so let's start here by looking at certification filters so by clicking all Windows computers basically you start by creating filter then you create a template the template has to reference the filter then you create an audit the audit has to reference reference the template so the filter is kind of what you would expect right computer operating system contains windows so that's your starting point and then the template will reference this filter so Windows computer minimum statistics this is when you'll actually start to hone in what specific things you're looking for what attributes you're looking for so you can see the filter here is referenced and you can see the condition Ram has to be greater than a thousand CPU count is greater than or is equal to one you can actually get deeper here too certification CI relationship conditions user relationship condition certification group relationships and related list conditions so you can get really detailed in here if you want to to build out your template you can build yeah pretty much your heart's content here there's a lot you can accomplish and then the last piece you'll go ahead and create the audit this is where the actual functionality it's kind of like a scheduled job right it's like cool you came up with your filter you have the template which refines what you actually want to search for let's go ahead and create a schedule and tell us what to do with this information so if I click in the audit here's a couple of good things to see here again this is where the template was referenced if I wanted to create tasks right this is where you do it Nifty little check box you know you can say group field specific group so you can assign it to a group which is probably what I would do if I did these that way it's not just one person you can run this on a schedule so just like any other scheduled task or scheduled job so daily weekly monthly periodically once on demand and then you have to run this audit so you run the audit and then you have to go back to the dashboard and then run that other scheduled job so if you have this guy running daily then you have the other one running daily it will update so keep in mind audits run independent of the scheduled jobs from the uh the dashboard itself but once you actually run one right then all these audit results start showing up so these are all the devices that basically didn't make the cut and I'll tell you why so Ram is not greater than a thousand if I did have a task this would also be showing up you click into it and it basically just takes all this data pops it into a description and assigns it to somebody and says hey remediate this so pretty useful I would definitely say get in there before you turn on any of these dashboard jobs before you go ahead and you enable the uh the compliance one maybe make sure that those audits are reviewed and turned off otherwise it's just going to start pumping out garbage data and I'm just saying Mark thanks for your comments I'd often recommend to set the compliance widget to have no influence on the overall seem to be Health scorecard while a useful tool the majority of use cases are not related to help the cmdb data yep that's probably true thanks for that so that's compliance in a nutshell as you saw you can get a lot deeper on that all right correctness this is where I think most of the value is going to be like achieved here so we're looking at things like duplicate Ci's orphan CIS and stale CIS and we're going to be looking at things like how do we configure and manage these results mostly um deduplication tasks right are going to be the thing where there's going to be manual intervention happening there you can also create workflows to handle those as well but there is some level of manual intervention where somebody has to go through and actually do something with that information orphan rules as well out of the box there are no orphan rules turned on so if you for instance created I think the example I have here is a web application you know is it running on a server if you had a relationship you knew that you absolutely wanted to track and you said If This Server is out here I want to make sure software is running on it um you know and and you want to make sure that it wasn't just stranded out by itself you want to make sure that there's a relationship between those you'd have to go configure those otherwise this just shows up as no data is available and then another important thing to mention is there's only one staleness rule to find out a box it's not going to really work for everything I'll again I'll I'll show you these guys here in just a minute but those are some of the high level notes all right and now we're looking at the correctness scorecard and there's actually you'll see some of the correctness stuff actually shows up here on the widgets the out of the box dashboard here duplicate CIS orphan CI stale CIS again I created these before this Linux one probably isn't the best use case this was one of those I think for this I set up said hey Linux server has to be in this specific computer room this email server has to be running on a server so I just went through and configured those guys so that there was some data here still CIS I haven't touched any of this information in here so out of the box you will get some staleness but maybe it doesn't make sense the out of the box Stillness you know rules for you might not make sense so let's take a dive into the orphan CIS and stale CIS first and then leave duplicate CIS for last because this one's a little bit more involved so going back to the CI class manager I'll look at email server and show you the orphan rule that I created all right and again going back under the health piece here we're going to click on correctness this time so duplicates is something that's handled by the system the ire it will do some checks that really you don't have a ton of control over the duplicates piece or I guess if you're really into the cmdb you can probably adjust it but it's based on identification rules and whatnot so that's it's pretty good at finding duplicates in my experience and you probably won't try to customize too much of that stuff anyway so orphan rule I will touch on that again clicking on the orphan rule here I just had a basic relationship so you can either do this based on attributes or a relationship again I'm going into a specific class and I'm setting the relationship value here so runs on runs on a server is all that this was so it's what it's going to do is when it runs this scheduled job it's going to look at it and say it's going to evaluate these relationships on the CI that fall under this class and if there is not a relationship where it runs on runs you know runs on a server essentially it's going to show up in Flag it and say why is this email server not running on a server and then you can look at it and try to remediate it from there again just to point out the scorecards are again they work the way that I pointed out the very first time we were looking at the completeness tab you basically are just setting your threshold there so let's actually look at one of these orphan CIS here so I just clicked on this right here and the data that we're looking at that pops up here's the health result if you're seeing a pattern here that's because this is how this works cmdb health results they show you what you're looking for and they pull in some basic information like Ci class name when it was evaluated and then a basic description so all that you're going to get from this is CI relationships do not satisfy the relationship rule again if you want to go through and assign a task to somebody you can my kind of rule here is if you know for sure you have a relationship you want to see standardized across the board find your golden child that was discovered and reference that rule make that rule based on how that worked on that relationship this is one of those somebody's going to have to have a decent amount of knowledge about this to to use this effectively but it can show you stragg layers or maybe places where relationships aren't happening the way that you would hope for looking at stale CI is next if I want to look at the staleness rule again like I said there's only the one out of box once it loads yep it's every 60 days so if a CI doesn't check in for 60 days it just applies everywhere across the board well let's just say email servers maybe this needs to check in a little bit more often you just click new and there there are some too that right like maybe 60 days a lot maybe you want it to be 45 days again that's on the CI level so you may want to change the out of box one you can go through each individual class and you can specify so effective duration if I just said every day right which might be a little excessive but it is an email server especially if you add just like one main one well you'd probably want that in there so either way you can change this this is where you would do this so you can change days hours minutes seconds obviously what you would want to do is you would you would want to look at like I think Linux servers or like let's talk virtual machines right maybe if one hasn't been on for 60 days maybe they pulled that data back and it's no longer relevant so you want to figure out what to do with these types of things and kind of remediate those from there so that's your stay on this rule again that's configured on each class but the only one that's configured out of box is cmdbci so you can see how as I've been talking through this uh some of these numbers are going to be a little skewed when you look at it right this will be 100 this will not have any data in it if you deactivate all those out of the box Audits and you don't actually set up any orphans won't have any relationships in here and then the staleness is based on 60 days hasn't checked in yet so there's definitely some information in here that you can update Paul yeah he just brought up a good point would require you to run Discovery or service graph import every day if you're going to have one day stay on this rule that's correct so you would have to make sure I wouldn't probably suggest I have a one day still in this rule I think that was just an extreme example there that I did you would probably want to do you know set a threshold based on what how often something checks in based on your schedules and things like that all right so I said I was going to talk about duplicate CIS and indeed I will so again this is a kind of a deeper subject there is a lot of information out there on duplicate Ci's how to remediate them and proper steps to take I won't pretend to be a total expert on that process but I can show you some simple examples that you want to run through I will however quote Sandeep on this one that when you are actually trying to fix your duplicate CI problems fixing one offs is fine but what's actually more important is trying to find out where the cross-section of these issues is stemming from so if you keep on having the same relationship create the same duplicates over and over again fixing that problem is going to be a lot more valuable to the organization than going ahead and assigning a task to somebody to remediate these guys over and over you again so you kind of want to systematically evaluate these understand how these are being created find the root cause and then fix that problem otherwise it's just going to keep happening and though it will keep people busy it won't be a good use of people's time all right so I just clicked this right here and then we're going to look at the health results so you'll see several things just at a glance here so you have your CI you have your task your class name last evaluated on pretty standard rights the health results this is where it gets more useful though is when you actually click into this duplicate task and by the way before I go there if you actually type in duplicate here duplication tasks you can actually open them from here and work them there as well so you don't have to necessarily go through the dashboard to do it this way you can go there all right so one of these tasks what's it look like here duplicate records found in one class computer okay and it gives you some basic information here based on attributes specified in the identification rules for the Ci's class okay so again like I said identification rules are something that you need to know a little bit more about the cmdb to get involved in your Discovery folks are definitely like yep for sure anyway so all duplicate CIS have identif identical values for below Criterion attributes so you can see here that there are a lot of duplicates in here this would be a good use case for figuring out why this keeps happening try to evaluate the relationship and whatnot so there's a couple of ways of fixing these kinds of things step one pretty much to click remediate I would probably do this on a simpler example I guess one there where there's maybe not 141 dupes but if you click remediate it starts to pipe you through this duplicate CI remediator which is recommended or like I mentioned earlier you can use a custom remediation workflow here out of box there are none so if you knew you kept having this problem and maybe you didn't know how to fix it but you knew how to patch it or maybe you know some combination of the two you could actually create a workflow that runs here to actually address the problem so you can get more detailed otherwise you can use the duplicate CI remediator and we'll just walk through I'm definitely going to be interested in some feedback during this process because this is a this is a lot of Records to be looking at simultaneously here once you use this out of box duplicate resolver it pops you into this screen right and you can select your main CI so it tells you recommendations most related items old is created newest updated a lot of the stuff keys off of Discovery Source right so I want to say so the duplicates in the cmdb dashboard are derived from the CIS that are tagged with the discovery Source field of duplicate when this happens the duplication task records created listing or grouping all the duplicate CIS a CI is tagged as a duplicate either during service now Discovery identification phase the ire Hardware rules used to identify if the CI exists or doesn't exist and basically if an update or an insert should occur if multiple CIS exist and they're identified by the hardware rule then the oldest CI in terms of the creation date so you can see old is created right here is updated and its Discovery sources to find a service now if we start out here essentially what we have are some options I'm going to select my main CI I'm going to say most related items ah let's just choose this one right here and then we're going to go ahead and just choose to remediate manually so you can see some of these options that pop up here and it's going to walk you through all of these things so you can choose the merge attributes right so I can see the main CI value other values so there's a lot this is why I said this might not be the best example but this is going to be the kind of data you run into so if we just accept all of these right you're going to want to go through here and and modify what works for you it doesn't work but I'm just going to take this and say merge all relationships from duplicate Ci's to the main you can also say don't merge them or you can choose specific pieces of it so you can see it gets pretty detailed in here pretty quickly you can also set attributes to custom values here if you wanted to this is where you'd say oh say I want to update it to whatever the heck I wanted to like assign to I want it to be a specific person otherwise you can just go ahead and delete so this would delete all the duplicate Ci's along with the reference records and everything basically merging everything together deleting all the others you can see how easy this could be obviously you'll want to evaluate all of these first you would look through relationships and you know related items and all these things to make sure that you're not going to just screw things up completely and you can see it'll pop up all these so again you would want to review this whole process a little bit more thoroughly than I just did right there but then it will go ahead and it will run this so for each one of these deduplication tasks it will actually pop up and and give you these options so this was the correctness piece here again duplicates orphan stale CIS you can see how this would actually be useful there's the relationship Health which is the last piece of this I won't dive too deep into it mainly because there's a lot to it and a lot of this is not super configurable so again this is just looking at your relationships if you wanted to update any number of these things it is possible to to do just about anything obviously as we know in service now but again this is where you would come into the CI class manager and you can actually look at things like dependent relationships right and you can start to override dependencies I wouldn't suggest this unless you really knew what you were doing you can update suggested relationships in here so remember before I said the duplication stuff was looking at identification rules this is again this is all in the CI class manager if you don't know how to use this and you're in the cmdb I would certainly take time to get to know this guy right here and how you can use it because you can see just right here the health piece alone what we've been touching was very useful but you can also see all of these relationship rules right here depends on runs on connects to all the stuff configurable and very useful so there's a lot in there I kind of glanced over this piece honestly I haven't seen anybody use this super effectively I would be interested in feedback if somebody else has and maybe better use cases there's certain stuff you just can't configure in the relationship kpi health results you can configure the scorecards like I said everywhere else and then you can use the CI class manager to see your relationships outside of that stuff that you would take from this and move forward with right how do you address the challenge of keeping your cmdb healthy well you prevent bad data which is easier said than done obviously that I think it starts in Discovery I think it starts on you know it's all additive I think you can all see how I kind of talked through requiring certain fields and making sure data is coming in correctly um you know our model is correct and Etc when Ci's come in maybe normalization there's a lot you can do there to prevent bad data but running and reviewing the cmdb data reports regularly so running these scheduled jobs having somebody who actually understands it and how it works and then performing your data quality reviews right somebody who can get in there again and look at these tasks or they know really what's going on there choosing specific CI classes that you want to see applications and servers and virtual machines and whatever for the inclusion rules to make sure you're actually getting the data that you want to see and not a touch a ton of erroneous data determining required recommended Fields obviously that's kind of like putting a squeegee on everything before it even comes in creating staleness rules so you can figure out how to purge stuff that you know maybe set an updated State when it hasn't been discovered in a while so again you don't just have an egregious amount of Ci's that are that are old and your orphan rules as we kind of talked about things like applications being related to a server virtual servers should be related to Virtual machines VMS should be related to hypervisor web sphere servers should contain location ear file stuff like that right like where you'll actually get value out of it I'd like to open this now for questions and hopefully some answers Brad it's a simple question really but when you went over the thresholds there you know 33.67 yep orphan steel is that I'm often wondering I've ever seen from a client a real business case for a use case where those values would be set so you know I know you can set them there but what you know to present that to a client and just say you know we recommend this but for the following reasons you know that's something I haven't been clear on so be interested to keep up on that for sure you know I think if you had some infrastructure stuff right like maybe you set the threshold a little bit higher say that you don't want duplicates of servers or something I'm just tossing out an example you know maybe maybe you set that threshold a little bit higher you say Hey you know if if I have like a small number of these affected I want to know and I want to draw attention to it that's the only reason why you would do it it's just a visual aid there is no actual repercussion if you set that Threshold at 99 and turned it red all it's going to do is just turn it red so if something was more important to you right like uh I would say an orphan or staleness type of thing and say hey hold on we have these infrastructure servers are stale then you draw attention to that specifically I could see that being useful otherwise I I don't know I don't imagine too many people getting in there and touching that yeah I think uh Carl that's a good call out customers like a healthy cmdb but they don't like managing it that's a good point let me go ahead and just plug in Glide fast there and say well that's what we're here for we can help do that yeah setting up the rules don't automatically make and or keep your scene to be healthy I think that's that's absolutely spot on Carl people need to have a process that they follow and people struggle with process I think and especially when it comes down to doing more this a lot of Administrators might look at that and say so I have to do more now but really it's preventative and it will actually help save them money and I think if you can get Management on board with that that's actually the value you're going to get out of that so Jesse asked a question when you use the duplicate CI remediator does it update so you don't get duplicates the next time you run Discovery my guess is it will go ahead and keep doing it this is where you would have to either adjust the identification rules or figure out why those duplicates exist it might not be as many as there once were at the onset of fixing the duplicates especially if some of that data is really old right some of it could be like a manual import if somebody created a manually then Discovery found it and they became duplicates it should fix it I would definitely say it'll reduce it but it this also goes back to what I was saying that you need to identify the source of the issue and then fix that there otherwise I would assume that it's going to going to do it again yeah I agree with you know identifying The Source I was going to say that it kind of depends on a couple things are they only using Discovery or are they using like service graph connectors as well because those could bring in be bringing in duplicate data and depending on the version of servicenow and how how well they're using those service graph connectors and using ire you know all those things can come together or not come together and and and make duplicates happen so you know if they're only using Discovery and on Tokyo probably be in pretty good shape you know but if they're they have anything else coming in and in an older way then that could be a challenge absolutely cool great questions feel free to ping me if you guys have more questions on this it's definitely an area I've been working on recently and you know love answering questions and hearing feedback [Music] thank you [Music]