all right well it's 10:03 so we're going to go ahead and get started with the presentation today again thank you all for joining we're really thrilled that you're here to be with us today so before we get started here um we'll just have a couple of housekeeping items that we'll go through um so first of all um don't miss out on some of our other webinars we have a full slated Deep dive webinar scheduled for this year uh we'll be rerunning customization versus configuration which was one of our most popular ones last year as well as some new ones like um how to set up your reporting working with citizen development and one like this care and feeding of your instance so make sure you scan here or um grab the chat uh the link from the chat um to look at our schedule for the coming year and no presentation would be complete without the forward-looking statements slide so if there's any statements made here today that indicates product developments or future enhancements in the product they are not guarantees that those are actually going to happen all right let's get down to to the heart of the matter here what is the problem we're here to solve today so well first of all we're going to familiarize you with the idea of the fact that you have a service now estate I'm sure a lot of you probably didn't know you had an estate um but you do and what we consider that is all your instances clones and Integrations and this also includes all the data in your system and all the way your users can interact with it so as platform owners or other roles in the service now uh ecosystem you may be looking to optimize or prevent uh performance issues um obviously I think most of you know and why you're here is if you don't intervene and put things uh checks and balances into place that that will become a problem and also we need approach that is consistent and regularly managed so that's what the problem is that we're setting out to help you solve today um so here is our agenda that we'll be going through um so we'll start with the intros so you can kind of get to know each of us a tiny bit but we're not going to dwell on that because I know you want to get into the the heart of the matter here which is going through our instance table and data management sections so without any further Ado such a cliche um my name is bre Koke and I'm a senior marketing manager and I've been with service now for about 14 and a half years I spent 10 years in the field helping customers Implement service now and since then I've had some roles in leading practices and in the marketing team where we concentrate on post content marketing to do which is really trying to provide you with information you need to adopt um our platform so Gareth would you like to take it off next and give us a brief introduction absolutely I'm Gareth Milwood uh I'm in the product success team here uh which tries to help customers like you adopt product to the best of its ability uh previously I was also in leading practices and I've been implementing customers uh with mostly customer service systems for the last 20 OD years Jared hi I'm Jared Mt uh I'm on the platform product management team focusing on evangelism again helping everyone get the most out of their service now investment and last but not least Sean hi folks uh super excited to be here my name is sea mlan I'm a master technical trainer with service now previous as with uh Bren some of the other folks I've had experience implementing uh at a small consultancy before I joined service now almost 13 years ago um and been involved in implementations and mostly training delivery of training and those kinds of things great thank you everybody so before we get into it we're going to start with a poll because we want to try to make this interactive so you can interact with us today and make this a little more engaging not just talking to some slides so first question question is do you already have some kind of a Cadence for maintaining your instances um and so this is a single Choice um question here so no uh maybe yes but we just review before upgrades yes we have a Cadence but it's not great um or yes we we have and we're here to learn how to optimize and Sean I think you have the wrong PLL up Sean one SE sorry about that folks we'll take those answers anyway though you may it yeah you may get twice the chance there we go all right thank you sorry about that folks thanks for the responses Fast and Furious there I'm hoping not to see a lot of number ones but it looks like we got about 30% there um with not a regular Cadence so this webinar should be super helpful to you today as we say in Minnesota everything's super this or super that all right we'll give you a couple more seconds here to fill this out thank you for your patience as we get this sorted so it looks like it's pretty even across the board here um pretty interesting yeah that doesn't happen very often in these these polls no no not at all all right so I think we have something for everyone here all right so I'm going to move along here and we will get over to our first section which is instance management um so this is the highest level of you know where we're talking about managing your instances and since I am not the expert on this I'll be uh sending this over to Jared take it away hey thank you yeah I've been in the service now ecosystem for about 10 10 years now uh been in your shoes a a customer administrator uh been a developer and yeah one of the things that came up a lot um when I when I owned an instance was those occasional tickets of hey this is slow this is you know what's going on this is it's different today so overall um our recommendation is to uh routinely take a look at things um and so you you know when um parts of the platform are are diverging from their normal operating Behavior Uh so yeah we'll want to review configuration of course but also that second bullet point uh taking a look at logs and errors that way I mean I my instances were never perfectly clean I never got it down to zero warnings and zero errors but you kind of if you look at it regularly you kind of get a feel for what normal operating is and then we'll go a lot deeper into this maintain your tables for PE for performance later on with some outerbox tools um but it's not just it's not just take a look um on a regular basis we uh can go on to that next slide and talk about some more specifics um on on a uh on a particular Cadence and one thing I want to bring up just before we go to the next slide and this is not a sales pitch but um if you already do own impact Advanced or total there you do have access to the impact uh instance Observer so you may want to talk to your service now team about leveraging that to help you with some of these tasks um and so now I'm going to turn over to Sean here where he's gonna um take us through a little exercise here on um what your Cadence find out what you're currently doing yeah and and um one of the things I wanted to do as we were talking through these pieces is to try to pull some of those pieces from you with um I'm looking with more than a 100 people in this room you can contribute as much as as we can to hear some really unique insights so I was looking to see if you could put in the chat what are some of the things that are on your administrative speaking of cadence you know what's on your administrative daily weekly monthly and quarterly lists and I'm I'm I'm looking for call outs to see on the chat for what you have there are some great documents out there and also some great conversations in the community on some recommendations for that some led by Mark Stanger from 10 years ago that are re uh brought up again as recently as a few months ago but I want to hear or see what do we have in the chat for what's your day-to-day check checklist what do you do every day in the instance for maintenance don't be shy folks there's no wrong answers here and if it's nothing you can also say you're not doing anything right now so all right everybody's being shy come on let's see some uh responses here all right nothing there we go thank you security nothing I do have a there was one that was HR offboarding which is a daily uh for info security that's that's smart to be checking and get those people out um Miguel points out reviewing mid servers um and that's a that makes sense as a pretty daily and regular check I see some security items in there um we do have uh a deeper dive on security plan for later this year so stay tuned to our our series oh I'm like going wait I want that um one of the other ones I got some people that have them as questions and I would go they're not questions they're perfect some people are commented on uh checking logs you know reviewing system logs and system Diagnostics is one of the biggest and easiest to kind of put into your daily schedule in fact some people I've worked with uh that administrate a system and it sounds ironic but they keep a log of the daily checks of their logs or they'll have a checklist going I did this and these are the standouts to me um yeah some of the other ones that are coming through and again I think the part of the reason I particularly wanted to do the Cadence piece by getting your feedback on it was you would get things that we might not suggest we can look at the daily and say review system Diagnostics or review the previous uh transaction log uh to see you know status and overhead but some of the ones that you coming through looking for data errors or checking uh Discovery schedules so several of you brought up those things those might not be things that you know the the four of us on the panel might have thrown into the list and you can look back at the webinar chat and go oh hey here's one that we should be doing over in our or so we got a lot of dailies how about weeklies us just with heavy data sets oh here's a domain domain set instances yeah so what are some of the weekly ones or maybe some of those dailies were weekly or other Cadence um what I should do in here and I'll make a point is just to separate it okay chat I put weekly so we can see the ones below that scheduled per requirements daily weekly monthly quarterly security findings performance status some people mention log and I think what's kind of interesting is we can take those same log files and look for you know in the log data be looking for large log files or overly sized at the end of you know like that's more of a semi-frequently like weekly would be a good Cadence to run those um or look at slow queries or slow running performance jobs to see and identify so we can tailor those on the daily that's almost too frequently but weekly it makes more sense to kind of now I can track this for quarterly um mid server CPU memory and usage trending okay well of course if we did daily weekly uh let's talk about Cadence for monthly uh and let me put up my yeah and uh one of the things I'll do with the feedback we get from here I'm sorry there's there will be sort of like a kit at the end of the presentation that'll get emailed out to you and I'll take this information and we'll compile that into the the kit that we'll send out and also make it available on our new customer onboarding uh section on the community right and I want to call this out into the chat because one other conversation pieces that pops up David uh Perez points out we get millions of ss log entries but I'd like to understand what to do with those one of the great things we can do for even in what we've talked about daily weekly monthly is we can create with those SS log entries filters or schedules for them and if we create those filters that gives us the opportunity to generate reports that we run or look at weekly or monthly hit this filter to look at these millions of Vlogs filtered for these core things um anything that is over a duration of gives us our slow queries um and so we probably and and this is one of those things that when thinking about cadenced U maintenance we can create a system of queries that we can you know at the end of the day I run this at the end of the week I check this query log to see what's showing up um for and you know again in the filter it's easy enough to say you know uh for the past seven days um to not therefore I'm not looking at something that's in the millions I'm looking for something that would get me uh compiling the millions and grouping those uh results into the monthly sections gives me a little more breakout of things all right so we've got some great answers on The Daily weeklys monthlies uh and that of course leads us to quarterlies now so I mentioned in weeklys what we might be starting to do is taking those same logs and applying filters to them we could do the same thing in monthlies um in those monthly angles I might be looking at some things like starting to track table growth to make sure that in particular as we ramp new uh applications or as we make new applications available to wider audiences in our instance you know I remember doing this for an organization where we were bringing on um even in the itm suite we were bringing on change management to different departments so we wanted to be able to measure and check to make sure we didn't suddenly get like we did with incident where it you know it went from zero to 1,000 miles an hour like that um as we look at some of the quarterly ones we see the syncing uh licensing TR up which is a good one that's awesome yeah I love that Health request all right okay in the essence of time I think we should probably get moving on here so the next thing we want to talk a little bit about is instance consolidation and I think Gareth was going to take this one yeah so um if you want to reveal the words there BR uh I'm Gonna Keep Them secret um really we need you to ask yourself which environments have we got and why are they there uh it's very easy for sub production instances to accumulate and sometimes there just longer needed so for each of the sub production instances you've got just work out why they're there does it have a defined purpose is it being used and then you can look at do we really need them or could we consolidate them because in the end the more instances you have the more instances you have to manage and all of that management it just slowly eats up all the time so look at those instances see if they need different kinds of data and whether they could be joined together and we can look at an example of that um on the next slide we've got an example where maybe it's a rather complex uh environment we're starting with where you've got 10 instances not including production uh development test integration testing acceptance testing I won't read through the list you can read for yourselves uh but the purpose here is to think about how you can take those instances and bring them together and reduce that maintenance overhead so you could bring test and integration testing together because they really require the same kind of data they got the same requirements uh similarly uat and um training could be put together because again they really need uh similar kinds of data which is very reflective of your production environment likewise operational testing performance testing it's not quite the same as uat but they again they need similar kinds of data and the different kinds of testing you do there could easily be scheduled to work out different times and just ruce that Main overhead and of course there are some instances you really do need to keep separate Dev you always need a Dev environment to be able to make those changes administrators might want somewhere to have a play around and try new features see whether it's going to work for them and if you're testing a new upgrade definitely you want a different uh instance to test that new version and maybe you've got some destructive testing that you've done and you just don't need anymore just delete it get rid of it once it's no longer needed and you will make everyone's lives that little bit easier yeah no kidding um so uh I think that brings us to our next topic which is system cloning guidance and kid considerations now that we've talked about what kind of instances how you could consolidate your instances absolutely uh so obviously you're not going to want to rebuild all of your instances separately there's a system clone functionality to to manage those sub production instances and keep them up to date so that basically works by taking a copy of the production backup data and putting into your subu production instance uh when you're doing that and you are creating these clones talk to the people using it make sure they're kept up toate with what's going on your developers won't thank you if you go and refresh the dev system and Destroy some of the work that they've spent hours days possibly weeks working on uh at least give them the opportunity to back up their data if not say no don't refresh it just yet we're in the middle of something um you should also kind of make consider decision about what data you need in each of those instances it's very rare that your developers are going to need a full copy of everything in production whilst uat is going to need something that is very reflective of production because you want to make sure that those changes really work with the production style data set um all of this can be brought together into a clone profile you can use that clone profile to decide what data you want to preserve in the sub production environment which you might want want to preserve development data or previously created test data you can Define exclude tables which is data you definitely don't want to copy from production for example you don't want to copy customer data into development and you can Define some cleanup scripts to run after the Clone has happened you can use that to either recreate test data in a known state to modify or delete data if there's stuff you want to redact or uh manipulate to make it safe for use in a develop or testing environment and as you go along you can start to automate these steps you can set a schedule so that uat for example is refreshed every month so you know it's always kept up to date you can build those cleanup scripts so it makes it easier to make sure you always have a automated repeatable process and people know what to expect again the more you can automate the easier it is for everyone yeah one of my favorite purposes of those uh post phone scripts is uh we in prod had a a group called uh Dev instance admins and so after the Clone was done we'd have a script that added everyone in that group admin so the just one less thing for the the CIS admin to remember to do after they got that clone is complete email yeah one of my favorites was just a cleanup script which changed the color of the different environments so you could tell which one you were in instantly yes that color coding is so so good um all right great overview of the system clones um of course you know me I got another Poll for you so um do you feel like you have the expertise on your team to perform the maintenance based on what you're learning so far um so you know no we could use some more training yes but we could still use some more training or yes we we feel like we're confident and good to go so it looks like we're trending in the middle here that's surpris I've never heard people say we never need any more training and luckily for everyone on the call Sean has provided a list of training courses in the Authentics that we will share with you after the call um so yeah looks pretty pretty much there in the middle um couple more responses coming in here um well the cool thing about the training resources is there's a combination of both um like the instructor-led facilitated models where on the one hand you're you're there at a certain time which I know gets challenging to be there but you have the value of like we do now everybody able to share information and ideas and that training for system admin advance but there's a number of classes in those resources that are on demand at your own pace and that's kind of a a helpful thing too absolutely it looks like we've got pretty much everybody that's going to participate in here so I'll move on to the next slide here and our next section which is table management um so we're going to tackle this topic um about you know how you're going to manage your table growth as you start to put more data into the system and so I'm going to tee that up for our expert Jared to bring us some information on that yeah thank you yeah this is uh the first of three uh tools or or built-in products um to the platform form that we can use to manage some of those unruly tables uh I will shout out that these are just a few more tools for your toolkit these aren't necessarily things that you're going to be using on a daily basis or even an annual basis uh some of these I've I've run across when I was debugging other things but uh personally have never needed to to configure them it's it uh in a lot of cases these are just already configured um out of box um and this is great uh yeah the big bonus of this is that these exist um so if you do run into one of these use cases you won't have to code it yourself or build a custom flow uh these are fully supported outof boox tools that have been around for a while and and very stable so the first one up is table rotation uh so um again this doesn't apply to every table uh so not every app will will need this but for those extra large um tables where the data that the most important part of here is things that are uh only required to be held on the platform for a short period of time uh these these are not your incidents these are not you know approvals things like that that you'd want to have for accountability purposes in some cases for years these are things that uh are are generated quickly and kind of go away um and let's see there they're broken up into these shards um so if you're looking at the table manager behind the scenes you'll see things where it'll be um CIS log o1 will be the actual table name but you'll access it by sis log and I don't know any uh like I mentioned the uh if if anyone else in the group wants to jump in here um personally I I if you uh have a rare case because this is stored in multiple actual tables in the database if you need to change the schema of of any of these things um it it isn't automatically replicated to all T all the shards in this group so there is a um UI action if you tweak something you can click synchronize shards and it will synchronize it across all of the tables but for the most part it I've run into this but I've never had to add or configure this okay all right well let's talk about um some other things here that we need to address like table extension yeah this is kind of like the the cousin feature to the shards uh this is for things that we do want to keep around forever but um we don't necessarily want to have let's say there's let's say there's 50 million rows in this in this group of tables every time um the service desk pulls up a table every time the business office pulls up the table we don't want to search through everything we we kind of operationally are referencing the recent stuff but for regulatory purposes we need to keep everything so uh this is a a different model where behind the scenes we're creating still creating shards you'll still see that uh table Name o1 Behind the Scenes but uh you'll still just do all your development and reference it at uh as that primary table name and yeah that's did I move ahead to but the yeah the again these these are things where there's a lot of records and um even with database indexes and things like that um it's still the amount of IO input output that's referencing these tables uh the indexes are are not enough to keep it under check so um we have these further options and again you don't have to copy down this this particular slide uh this is just letting you know these things exist uh when you do run into this in the real world um our docs website goes into good detail about all of these the finder points of uh the the how we want to make sure we're querying based off of the created on date and things like that so just when you're walking around taking a look under the hood in your instance um if you see something like yeah like that sis logor o22 you'll know that you're looking at a special type of table group but is this like if you see something say something at the airport it is this is yeah don't yeah don't don't get uh don't get freaked out that um this this explains what you are looking at um and so those are those are the two types um again not for every app not for every table but for those special use cases and you don't want to do a hack or or build something yourself these are stable outof boox platform supported uh methods and the third one we have is probably the most most common of the three that we'll talk about this is the table cleaner uh the table cleaner is not eligible to be used on those table rotation or table extension The Shard models uh so this this can be used on on just about everything else um and what we do is uh instead of again instead of creating a custom flow or custom script to move some of those uh things um to purge those those data we have a a clean outof boox method for you to do that and let's see a good time saer for people then they don't have to create the script and do all that themselves yes um my my favorite feature of the table cleaner is I'm not sure if it goes into it in the next slide um Yes actually actually it does over over uh in the don't column here don't expect a single run to remove large volumes so this is very performance aware um it will uh run in the background um I believe it's if if we look under the hood it's going to run once an hour uh and is very lightweight so in a single run it will run for a period of time and then we'll pause uh and catch up on everything else later so if you happen to hit a particular Milestone where um a million records need to be cleaned uh don't expect it all to be done in one hour we don't want to hammer the database we want to kind of chip away at it hour over hour and everything will get get caught up so um so yeah and I'm seeing in the chat we're asking for some specific examples of a table cleaner um so I know this is a little bit different than archiving so table cleaning we're just we're just getting rid of things um anyone else in the group want to jump in with um sorry what was that Gareth did you have one I was uh no I realized I was wrong actually I was about to say the table cleaner can be used clean up archives but we actually got destroy rules for that um uh the the CIS log on for example the uh sharding for sharding um table rotation is already defined for um CIS log so CIS log will be around for a period of time and then it will age out as that data gets overr written whereas another example of sharding would be uh the audit logs that's extension instead so the audit logs will just keep extending and will keep going on forever until you are serviced now to come and remove some of them which is a much more Awkward process but the whole point about audit logs is you don't need to get rid of them yeah and Harrison one of the other things I wanted to throw out to that question of you know real examples and real context first anybody in the in the participants that that has it to contribute but second um the just as much as we check in the docs I'd really go as well to the community because in the community I and again it's I think one of the references I pointed in the list that we'll be sharing later but uh some of the community discussions on this point to um organizations that are using table cleaner for X versus using it for y and I feel like that's kind of almost a a really better than you know because you're hearing the feedback from many customers uh because they'll all be jumping on that Community thread to go yeah this or not that um and like I said I sent one link but I'm I'm not not sure on the one for table cleaner specifically yeah I mean I think Doc is really good for telling you how to do things and the community is really good for giving you some context or a color on yeah that's kind of the way I think of it like Doc is the official but Community is the real world we've done this or what about you know or to see people who've walked the walk so to speak as I know customers always want to know what other customers did um that's you know it's always helpful information right yeah you threw out uh Mark the name Mark Stanger from the the old school Community uh all through out uh and I think he's even in attendance in the in the group today the new school Mark is Mark roof um he's one of the European Dev MVPs and uh not so much for new customer instances but some of those instances that have been around for a decade or so uh the database sizes get large with hidden attachments and things like that and recently he's had some really great content on how to detect that and how to purge that um so new Mark uh thanks for joining um I I will say that back to the table cleaner the example I have is maybe you've seen this in your organizations's uh email policy where we have data retention we say depending on what industry we're in let's say we must retain things for two years but then we absolutely just have to delete it after two years right because we don't want to just keep everything forever for legal reasons for governance reasons whatever the case may be so um within the context of service now there could be depending on your organization's requirements uh inbound emails we want to we absolutely have to keep those for two you know two years or we um incidents or whatever the you know whatever the record type happens to be but then we have to purge it and we have to audit the fact that we are Pur in it and so this would be one of those tools that you could use to to make sure that we're not keeping things around and you know what Jared that is a perfect lead in to our next topic excellent way to go that wasn't planned by the way but it just worked out really well so um our next topic is data management um which you know I think Jared was referring to some of the challenges around that and obviously you know one of the first things you need to understand is what are your retention policies for your business for that record type um and it's just as dangerous to keep a record for too long as to not have it um I've had experiences when I was in pharmaceutical industry where you know we didn't hadn't gotten rid of some records and then you know you get sued and they do Discovery and that information is available and it really shouldn't have been available so there there pren you know there's dangers of either side of that retention so it's really important to know um what your document types are and and what your policies are and of course because um we're very thorough about these things we have a dos and don'ts slide for this as well um so just some basic things here like records should be readon you know obviously they should be searchable and easily defined access should be limited and not everybody gets to have access um and the ability to audit and notifications of any non-compliance and then obviously you want to make sure you track those expiration dates so you get that data out of there before it becomes a liability um and then make sure you have a backup um you know as a Somebody I Used to Know used to say one is none to is one um so think keep that in mind with you know these important records and then obviously this is just a you know General best practice you don't want to print out those um records because then you have an uncontrollable version of that that may or may not get destroyed on the appropriate date or may get into the wrong hands well that's just a quick uh overview of retention to kind kind of give you some things to think about and make sure you have those on your radar and I'm going to turn it over to Sean now and he's going to talk about some of the tools that we have available to help with some of these retention challenges oh well there are a couple of different things available and and some of them we've we've already kind of hinted at archiving is uh one of those pieces that we can use to offload old data um we've already talked about being able to um run archive and archive and Purge but the other ones that are out there including CD data manager and email retention are plugins uh and it should be mentioned that in particular cmdb uh data manager is one of those plugins that I think with Vancouver uh it started in Rome um and then I think as a Vancouver is uh turned on by default uh on the other hand I was playing with an instance this morning that is a Vancouver uh for one of the classes they teach and I found I was having to find the plugin to set it up and run it so that's one of those pieces that you might want to look into more so that we know in terms of that tools on the governance side of things in particular with the data man uh cmdb data manager uh there was a question about that as well so I want to clarify the cmdb data manager is not all data it's configuration management and it's particularly tied to creating a governance policy around um when can I what rules and what filters do I take about making configuration items um ready for retirement or then eventually dispersal so if I'm if I'm doing that then what I'm doing in the tool is setting filters and adjusting the filters accordingly based on my U organization or um Fields governance and policies email retention similarly a a policy plug-in that you can download and as Aaron pointed out we want to be careful because some of them it's a it's an either or rule right there's a I want to retain it for this long but longer than a certain period has issues or risks as well and um both of those tools in particular when you have the plugin are essentially identifying classes uh and filtering uh based on the you know in configuration based on these classes of CI based on these rules and this potential approval then we can take these next actions to remove them through um and so that's what I wanted to mention when we talked I I when we were discussing these slides as a group I was mentioning the logs and other management um just as a still wanting it out there not so much that there are tools directly towards logs other than the ones that uh I believe Gareth mentioned uh with regard to the system log and some of the other logs that you know we we go through and have cleaning but that should be kept in mind from a governance perspective as well um because those those logs that build up over time we want to make sure that we're fully removing them when audit and compliance conditions come into play so I think those are the pieces that I wanted to mention about those tools and to see what else maybe my colleagues can throw in oh sure sorry I went to jump the gun here no that's perfect do and don'ts yes I think Gareth was gonna take us through I'll talk about this one um so archiving is really important um a while ago one of my colleagues was uh talking to me and he gave me a rule of thumb which stuck with me you shouldn't be putting data into a system unless you know when it's going to leave the system you should be defining your archive rules as part of your development you need to know what your life cycle is for the data um when you defining archive rules um you basically you build up criteria of things that you want to Archive and it will move that data into an archive table so it's still available in the system you can still access it if you need it but it's not going to be clocking up those operational t tables slowing down queries and making the jobs of your agents harder and slower um starts small uh and work up to your final archiving criteria if you are doing this in a relatively mature system uh and every time you create an archive rule you will almost always want to create a destroy rule as well so that you get rid of stuff uh once you no longer need it don't try and boil the ocean don't try and do it all at once don't try and archive uh the sharded tables anything which has got a rotation on extension on archiving isn't going to work and don't try and stack them you can only have one archive r on each table hopefully that makes sense to everyone um the next part yeah that leads us into our next topic of large attachment cleanup yeah uh large attachments are another way that um a lot of space can be consumed very quickly um there's no magic bullet for this one uh almost all customers need file attachments um customers need to talk to you about things employees need to show you how things are appearing there's all kinds of reasons for them uh I would encourage you to make some deliberate decisions about what kind of attachments you want and why you want them uh how large are they going to be and just do some planning decide on a a sensible maximum size limit and review it now and then what seem sensible for a mobile phone photo 5 years ago is not the same size as it is these days so you do need to keep these things under review regularly review how you're using attachments and don't try and use us as a file distribution platform service now is not good for that if you want to store multi- gigabyte files there are probably better places to do it and then we can work with them rather than trying to take that uh piece of functionality 100% yeah I keep hearing in my head that famous quote from uh somebody from just a few years ago who said we'll never need more than 64k of ram was more than a few years ago Sean yeah well um but to to gareth's point the other piece that I wanted to to comment with regard to large attachments is there's it's this sounds weird but it's never too late to iterate uh we can look at any uh area where we have files that are maybe large attachments and look to going is there we we could re-engineer add a field that has a link to um this system that does keep large files is there a way that we could go forward and make that adjustment so that I had a customer that was actually doing a lot of that they were putting um then in particular this came into the governance issue as well because they're putting Legal Information in and therefore keeping all their legal briefs and docks in there and that was like a let's do a reference to that file which then allows us to manage the security of it as well as not owning the those huge files of data so all right um CIS audit cleanup uh this is a a tricky one um I've done lots of implementations for customers and when you say to uh a lot of people in the business what do you need to audit uh the immediate response is everything and if you audit everything that that takes up a lot of space really quickly um my recommendation is think about what you actually need to audit and why there are going to be compliance aspects of your system that absolutely want to audit there are security aspects that you want to audit but you don't need to know the complete history of the title of a knowledge article going back to the dawn of time it's it's really not that important so again I keep saying make considered decisions I think um it's really key uh work out what tables you want to audit and then within those tables Are there specific Fields where you need field level auditing and make a deliberate Choice rather than just trying to make a blanket decision that covers literally everything in the world ever yeah because what I found with customers you know they want to save everything whether it's tickets or audit files or whatever um it gets to be a data management Nightmare and then they usually can't find what they're looking for anyway or they never refer back to it so making those very conscious decisions I think are is really important and can I do a quick shout out for the log export service um that that came out in uh Utah I believe it was generally available and lets you stream near real time out to your other corporate uh logging systems uh again to get that stuff Consolidated with your other Enterprise systems and then also we don't need to keep it in service now since the security team has access to those uh those things so there's a there's a free starter tier for that as well right help us on the way can you put the name of that in the chat just so everybody can see that that would be great oh Sean I just threw the docs in oh Sean look at you well I saw I as soon as I heard it I went oh wait let me look at this see we learned something new every day about service now the platform has so many capabilities and features yeah so next up we got some conversation here around the CIS log yeah um it's it's a different conversation from the audit U log because CIS log is rotated by default it's already going to disappear after a while but that doesn't stop it becoming a problem um particularly if you are developing your own customizations your own scripts um I've been a developer uh I've had to troubleshoot things by adding things to logs and it's very easy to solve the problem and go great it's done I move on and onto the next feature leaving those logs in place uh that's probably not a problem while you're in development it's probably not a problem when you get in test once you get into production and things scale it becomes thousands and T tens of thousands of log entries and it just makes finding other things in the logs really difficult uh so I by all means log logging is important but log exceptions not successes make sure that troubleshooting and debug logs actually get turned off after you have uh solved the problem and got into uat and make that aart of your day-to-day development cycle uh it can you do need them but you don't need to overly log right and um so in the appendix we're providing a diagram that will help you figure out what tools you can use to accomplish some of these tasks but this is sort of like a an overview like as we say the 8020 rule it might look something like this and I think Sean you were going to give us a little more color on this uh yikes I'm jumping to the slide hang on one sec on [Music] the I'm looking and going wait where are my notes where are my notes where are my notes um well I think mostly with with regard to building out those or prioritizing those um you know if we're if we're looking for managing the archiving piece is the most important piece that we want to look to because that's the bulk of what people are hitting right um and it's those tasks that people are are doing work with that we want to streamline the performance of the most so that makes sense to be first targets for uh managing and and trimming things out for similarly as we start looking to the other big pieces in terms of which do I Target iteratively um working out things like email as uh Jared mentioned um we need to be careful about holding things too long because of those risks as well as size cmdb when we look at the data manager we're looking at that again for accuracy and use so if if we kind of look at all of these components in terms of which ones do I tackle when um and we were to prioritize or set some criteria um we're looking for who's using what most and that's our first Target for what to manage most with archiving and and and just data management and then what is most important from govern governance is the next thing on our I think our tackle or table that we're structuring things like uh table rotation and as we were talking about large uh large data pieces can be iteratively improved um and so we should schedule in our time for Cadence to say you know maybe quarterly we're reviewing and going what things can we improve on all of this stuff that we've tackled so far all right well that pretty much wraps up our presentation for today we are going to run that poll number three again apologize for making you answer this question twice um but now that you've attended the webinar are you feel like you can explain to your management why you need to devote time and resources to this um care and feeding of your instances are you motivated to create a maintenance Cadence or you on the right track and definitely going to level up and this is a multiple choice so you can select more than one of these if you'd like um and I see the responses are coming in here fast and furious and also if you have a question please enter into the Q&A we'll start picking off those questions once we get through the poll here uh we'll start answering those questions live all right it looks like we've got a pretty good uh response here on the poll so let's go ahead and stop the poll Sean and we'll move on to the next thing um this is what we put in at the slide in every deck that we do with these webinars it's the summer or the too long didn't read so if you want to jump right to this slide you can come here and get the overview of what we're doing this is sort of for people who want to take a screenshot so they can remember what this was about and if they want to come back to the side the information later you can come back to it but this is just kind of a quick summary for you all right well I'm sorry folks we are at the top of the hour so that concludes our webinar for today I want to thank you all for joining um there will be a post event survey coming out to you please put your comments and suggestions in there we always iterating to try to make our webinars better and be on topic and um hope you all have a great rest of your day thanks for coming thanks folks thanks for including me guys