 Good morning. Good afternoon. Good evening. Welcome to another episode of Red Hat Enterprise. Linux presents. I am Chris Short, executive producer of OpenShift TV. I am joined by some of my favorite Red Hatters, Scott McBrien and Bob Davis. Today we are going to be talking about RHEL Workstation, which I think is awesome because I've used it before and its intended capacity and have had great success with it in a previous life. So Scott, how you doing today, buddy? I'm great, Chris. And as you mentioned, Bob Davis is with us. He's the principal product manager for Red Hat Enterprise Linux Workstation. So I'm going to ask Bob all the workstation questions. Thank you. Bob, over to you. So go ahead. Yeah, I was just going to say, you know, hi, everybody. My name is Bob. I've been with Red Hat for four and a half years or so, working in various capacities and product management. And now I run the RHEL Workstation offering. So I think a lot about how the people who actually sit at keyboards and in front of monitors or with their Cintiq tablets and styluses, how everybody uses RHEL Workstation to do the things that they do. So I think probably the first question is, what is RHEL Workstation and how is it different than RHEL? Well, RHEL is lots of things to do with different people. Workstation is, like I said, that's something that people actually interact with. So I tend to look at workstation as that confluence point between human creativity and high performance computing power. So it is where the digital artists can bring their character to life using, you know, all the wonderful software that people build for RHEL to make 3D animations, or it's where engineers work to make the cars that we drive safer, or where scientists sit and analyze data that comes off of super computers to develop medicines or understand some, you know, fundamental aspects of the universe. It's where some magical things happen in front of these screens. So it's really, really fun to talk to my customers about it. I bet. I mean, some of the, to give you some examples, right, like we're looking at large, you know, imagery data sets is some of my use cases from my past, some civil engineering use cases, right, bridges, dams, that kind of thing. And then I think most interesting, the most interesting use case I have is I had a bunch of friends that worked at a pharmaceutical company, and they were just building new medicines and like, sequencing stuff and doing all this crazy work. And this tiny, they had this little lab in the back of the building, and they all had these mega rig workstations to like literally build new medicines for people. And that I thought was the coolest use of workstation, because that's awesome. Yeah, it's it. People are doing really, really wonderful things. And I love that the I love the human component to it. Because so much of the time you think about rel as, you know, the operating system that makes all the internet work. And it's, you know, it's, it's sitting in those cold, loud rooms that have, you know, so a whole bunch of power running into it. But nobody ever sees it. Nobody ever touches it. Yeah. But this is, you know, this is where the rubber meets the road as it is. Yeah. Yeah. It used to be, though, that like workstation was a different variant of rail, right, than other rails. Yeah. And in fact, that really that change happened. So there used to be what we refer to as content differences. So there were packages that were included in server that are not included in rail workstation and vice versa. You don't need the, you know, a massive list of fonts, for example, in a server, generally speaking, that you would need in a workstation. You don't, you know, there were limitations around what you could do as far as using a workstation as a hypervisor or for you know, doing anything with containers. But with the launch of rail eight, we shifted over to a flat content set so that the content that you get in the server is the same as what you have in a workstation, which is fantastic because what it allows, say, a software developer to do is build and test software in an environment that they know is going to mirror what the production run or what the production environment is going to be. You know, you're not going to accidentally trip on some package that was included or not included in your production environment. You know, it's the same for, you know, let's say you're doing a, you know, 3D visual effects work on your workstation and then you ship it off to the render farm to be composited with all the lighting and the environments and everything else that has to happen to actually make a frame of of a, you know, a movie. Again, having everything be the same in your design and in your design environment as your render environment is important. So, you know, flattening that out has made a different, a good difference for us it makes the, when it comes to like what, what version should I buy, you can actually look at now, what am I using it for instead of what packages do I need to make sure are there, which is, you know, just a more customer friendly, I guess you could say, way of, way of looking at things. It's like, are people touching it or not? So, Scott, there's a question in chat. Is rel for devs valid for one year or is it just valid forever kind of thing? I'm a little confused. The individual, so, unfortunately, I used to work, my prior job was the, I was the rel experience product management person in the developer business unit. So, that subscription is a one year subscription that just on an annual basis when you find out that it's expired, now nothing stops. Right. Like it's not, you're not going to try to like boot rel one day and it's going to be like your subscription expired. Goodbye. Insert coin here. It doesn't, that doesn't happen. You just need to go in and you just need to go in and like check that everything's still the same and and re-agree to the terms and conditions for another year. The snag is that you cannot do that preemptively and usually what you have to do is like try to download something. So, if you download, like, download the most recent version of rel and it will ask you, it'll ask you for the re-subscription. Perfect. Thank you. Yeah, and I think from our last developer for individual subscription actually sent me an email telling me that it was going to expire and tell me the date. So, and then if you log on to the Red Hat customer portal, which is access.redhat.com, in the subscriptions field, it will be listed and it'll have the date on which it needs to be reviewed and there will be warnings there too when it's about to expire. So, but it's about mentioning like rel doesn't stop working. What ends up happening is your machine's access to the content development network won't authenticate until you re-apply your subscription and then the machine's still registering and then it's authorized to download the content again. Right, yeah. Cool. Good to know. Thank you. But back to our earlier thing. So, Bob, one of the things I found most irritating when I was a systems administrator was like a lot of times the packages between workstation and server were, they were all there with one or two very small exceptions. But it's like in the rel for server distro, we packaged some things up as like optional RPMs, which were Libre Office and those more desktoppy apps. Whereas in the workstation, those were part of the standard workstation channel. And then you'd have to find the server supplementary channel to get some of the server side applications. Because of course, we read out like somebody on a workstation would never need a DHCP server. And then we find out that there's that one guy that needs a DHCP server and we have to deal with it. Right. So, yeah, like flying the content like makes sysadmin live so much easier. Yeah, absolutely. And it's, yeah, it's, that was the whole idea was, you know, we were, we were doing something that made sense to us from where we sit. But it added complexity for enough customers that, you know, our effort to make things more streamlined actually turned out to be injecting pain for some, for us, you know, even if it's a small subset of customers, right? It's, you know, like, why don't we just make it the same everywhere? So, from a, from a sysadmin perspective, realm machine is a realm machine is a realm machine. And you don't have to, you know, have some of these crazy runbooks to figure out how to, how to do an update on something. Well, there was one other change that happened with realm 8 as well, where we did away with one of the product types that I think probably would have fallen into your Ballywick at one point, the desktop. Yeah. Desktop was a, so desktop in some ways was redundant. And when we were looking at the ways it was being used most frequently, it was almost more like a, an edge device or, or what we would refer to now as a distributed computing device. So this would be like a, like a point of sale station, you know, like a, you know, a cash register or some kind of a, a kiosk kind of use case. But it was not optimized necessarily for those particular things. So, you know, there, there are problems that are unique to edge use cases that we weren't necessarily accommodating in the product development for, for desktop. So what we've done is I think we, you know, we eliminated desktop. We're now focusing in on the workstation experience, which, you know, as we've described isn't, it's not that point of sale thing. It's that scientific animators, engineering, you know, geological exploration type of workstation. And for these, you know, for an application where you're running, or for, you know, a, a use case where you're running a single application that depends on, you know, a handful of touchscreen types of interfaces or things like that. That's where, you know, using edge types of installations are, are going to work better for our customers. Nice. So are there like specific things someone should look for when deciding whether they should use workstation or server or some other type of role? Well, the biggest one is going to be the use case. So, you know, think about where is it going to run, meaning what kind of hardware is going to run on, what kind of environment is it going to be in, and what kinds of things are in the end user going to be doing with it. So, you know, typically when, for workstations, it, it, it's going to imply that this is going to be a heavy graphical use. Typically, it's going to be one, it's going to be a, well, it's always a single user environment, just by, by definition. It's a single user environment where you're going to have typically a very, like a beefy kind of workstation, which is, you know, so, you know, probably two processors, very expensive type of professional grade graphics cards, very high power. It's going to have multiple screens or large screens. The input devices will, you know, a lot of the ones that I'm seeing, typically include like the large professional grade Wacom tablets for a lot of use cases for other ones, say, you know, using it for a, you know, a scientific analysis type of use case, it won't have that. But GPU heavy workloads designed for an individual to sit and work at a screen with a, you know, mouse and keyboard versus a server is, you know, obviously going to be something that's nine times, most of the time it's going to be running headless. So it's not going to have, it's not going to have any, any gooey components necessarily. It'll, it'll be in a data center. We also do have use cases where people are running workstations in data centers with individuals and remoting in using some type of, you know, a VDI kind of setup. Which is an interesting option for, it's an interesting option for customers that have a workstation dedicated to particular computational types of analyses where, you know, the data and the, the data that they're using needs to be kept protected. So they don't want it necessary. They don't want, they don't want it outside of a, they don't want it outside of the data center. But they need people in remote areas to be able to access it, even, or, you know, from remote offices or, or now even with coronavirus, there are a lot of these, a lot of people who had been traditionally using their high powered workstations in an office are now able to remote in from their dining room tables. Yeah, like, I would see that often from my friends, right, when they worked remote, like, yeah, actually, like, not RDP, but yeah, VNCN through a VPN kind of deal. And I use it like a VDI kind of deal. But I'm just actually sitting at home. And it's, like, if someone walked by, they would see stuff going on my screen, you know, that kind of thing. Yeah, and there are some very high end solutions, you know, such as nice DCV or Teri D'Hici that, that are being used in conjunction with high powered workstations where, you know, very low latency response times on key presses or, or, or in stylus input is required. And it's funny because they, you know, they refer to it in terms of like, frames per second. So when you hear, when you hear them talking about these input responses, it's like, oh, okay. So we're talking about like high powered gaming tech is now is turning into high powered workstation tech. Right. Yeah. It's like the same stuff you would use for Bitcoin mining, you would find in a workstation these days, right? Like, not that I would recommend real for Bitcoin mining. No, please don't save the environment folks. Anyways. So yeah, you know, like, which one to choose is going to be really dependent on, you know, is this a, is this a server cluster that's serving up an application for remote use? You know, that could be, or is it, you know, we have HPC high performance compute, which is, you know, depends on a massively parallel computational model where you've got thousands or tens of thousands of individual nodes working simultaneously. But you know, the beautiful part is that you can, there, you usually would wind up buying, you know, you buy server for, for the application that it's running for the, you know, for the final application as it's being run, you can buy workstations or use the developer subscriptions for, and workstation with that, with, you know, for your software developers to be building on the same platform that you're deploying to, or you use workstation in conjunction with your HPC for the, you know, the modeling and, animation rigging composite lighting, and then you use your HPC nodes for your final render and, you know, so that people can enjoy a beautiful animated film. Right. I think it's interesting the top 500, you know, supercomputers in the world, how many in that top 10 are rail systems is remarkable to me. I think we've got three of the top five. Yeah, like it's, there's like, and the most recent ones are using rail, which is cool, right. Like it's not like some, you know, some old HPC system, like, you know, it's national laboratories, massive one, Japan, I think has one. Yep. Yeah, there's, they're all over the planet. But yeah, the one in Japan is unique in that it also runs on arm. Right. Like power efficiency, super computing type things are being thought of and put into play already out in the world. It's pretty cool. So Bob, what are some of the, what are some of the developments that you're working on to pull into the workstation experience for rail? Well, obviously it's a, it's largely iterative. A lot of the things that we that we work on are driven by specific customer requests that come about as the nature of work changes or the things are the problems that we need to solve change. So, you know, I don't have any major breakthroughs to report, but, you know, things like support for HDR monitors as an example, you know, that's the kind of thing that we would be working on as they become more and more important for for our end users. Continuing to align the packages that we ship as part of workstation with different standards organizations so that, you know, it's, we have different, there are different standards organizations that drive like reference platforms that whole communities of software development companies depend on. So making sure that the, that the operating system platform that we deliver is going to work with the various software packages that people depend on is one of the things that we do by continuing to have ongoing conversations with those with those reference platform organizations and a lot as well as with the different software vendors and different hardware vendors to make sure that everything works together so you don't have to, you know, have six different systems to run six different software solutions. Because they're all setups just slightly differently. So, you know, a lot of it is just making sure that the end result is simpler. I think that, you know, one of the big things that I think is important when it comes to simplicity and flexibility is actually a flat pack. So flat pack if you're not aware is it's similar to a container based, it's a container basically it is a container that our flat packs are our OCI compliant containers that you ship a desktop application in. The difference is that there's this flat pack runtime environment that allows you to do things like cut and paste and all that between these different container applications. So it enables a great deal of flexibility because if you look at a lot of the professional grade packages, a lot of the support forums are constantly is if you do quick searches, you'll find a lot of questions about how do we even get this installed and what tweaks do I need to make here what tweaks do I need to make there. If it's running in a container, you get the application along with all of the dependencies and all the configurations that need to be there. It's all self contained. So it takes that pain away from an individual who needs to just install some packages to be able to use. Oh, but like we've been talking about that a little bit. And as I'm glad that you pivoted to this topic because I was about to bring it up and was worried that you would like slap me. No. Flat pack is exciting. Flat pack I think is going to be something that will it will make, you know, as we commit more fully to flat pack, you know, by providing flat pack options for various applications that we ship along with rel and then, you know, trying to do what we can to provide assistance to software vendors who want to also ship us flat pack. It should make those in the lives for the software vendors much easier. And it should make the, you know, an individual end user of a workstation who needs this application or this other, you know, commercial or otherwise. It makes it a whole lot easier for them is, you know, and then for all the, for all those organizations that are using, you know, in house built bespoke applications who just want an easy way of distributing internal, you know, distributing the app internally without a massive support load just to get the things installed. You know, it's so much better. It's so much better. I'm, you know, as a workstation product management, you have to kind of be passionate about the end user experience. Exactly. And, you know, if I can make, if I can do something that helps an artist and engineer, a scientist get into that mental flow state where suddenly they realize that they're starving because it's three o'clock in the afternoon and they forgot to eat lunch because they were in the zone and they were doing the best work that they've done all week for all month or they cracked that huge problem and they're just so excited that they lost track of time. If the fact that my operating system didn't get in their way on, you know, on the path to that mental state helps, that's super exciting for me. Yeah. And to be honest, the Fedora box sitting just over my shoulder there, it's filled with everything's pretty much installed through Flatpak at this point, right? Yeah. It's just easier for me to script it. It's easier for me to, you know, answer the lies that if I want to, if that's a word, it's just like, you know, package name, off you go, right? So where we're starting, I think, to enterprise up that technology, right? So Flatpak is really good in that it solves some problems like somebody upgraded image magic on your box and now your app doesn't work anymore because it required the image magic library, but now it's been updated and the API is different and blah. So you could take with your Flatpak, those key libraries and include them in your Flatpak so that they're stationary, right? And the operating system libraries can get updated, but the one used by your application stays consistent. So that's like awesome. The flip side of that coin is we updated the image magic on the operating system for a reason. Maybe there was a CVE and we were applying that update to close that security thing. Well, now your app is still running with the consistent, not updated version of that library. So, you know, from. So yeah, I mean, but that's the challenge of making Flatpak enterprise ready, right? Right. So like if you look at our, if you look at the Red Hat container catalog, you see that they have freshness grades, all the containers have freshness grades. And you can look and see what, you know, what's, you know, has a CVE been addressed? What's the latest version? When was the last time it was touched? These are all, like a lot of this stuff happens automatically. And what's really cool about our container images is that if you say, you know, you grab, say, a Node.js base image, it has other libraries besides just Node.js image. Well, if there's a CVE that was in anything, anything that's in that container image, that container image gets updated. And it's not just looking for Node.js updates. So you don't wind up in a situation where, you know, you're, where you, how are using this Node.js base image, but only Node.js gets updated. You're not, you're not going to be stuck in that unfortunate situation. So what we need to do in our, in our efforts with making, you know, anytime we adopt a new technology, making it enterprise grade is a big part of the value that Red Hat adds. So, you know, how, as we distribute, as, as we look at container or rather at Flatpak, it'll be, you know, our obligation to make this enterprise ready, which means providing a management and support experience that is in line with what people expect today. You know, if you've been a, if you've been a Red Hat systems administrator for a long time, I don't want you to have to go back to school and completely retool because I'm changing one thing. You know, that's not fair. I want to make sure that the expectations that you have for what a Red Hat product means are met. And that means, you know, that things like automatically updating the Flatpak containers when there's, when there's a problem in any component in the, you know, in the container itself, CVE or otherwise, you know, that those things always get updated on time, making sure that, you know, whatever you use for, you know, the tools that we ship for management today will be able to manage that in a way that is expected by seasoned users. So, you know, that's, that's a great point, Scott. That's, you know, like Flatpak in the wild today in Fedora is great and it provides a, you know, it's really easy to see the benefit of Flatpak when you use it in Fedora. Now we need to take that and, and how to, you know, and make it enterprise-grade and in line with what a REL administrator would expect. And the parallel with the Red Hat container catalog, I think it's a really good one because like we do automatic rebuilds of those containers every six weeks by default. And if there's a, you know, critical or important CVE opened against any component, and by the way we track all the components, you can look at it in the container catalog of what actually went in there to build that container. So we know when that's been, yeah, so we know what's been put in there. So when a CVE is open against that, they produce a fix for it, it's a rebuild all the way across all of the assets that are affected by that, that package change. So like we've done it with other technologies and hopefully we will continue to carry on and do that in newer technologies as well. That's the goal. So Bob, what's, what's one thing that you see people do with REL all the time that you wish they would either not do or do differently? I love this question. So much pressure. So much pressure. You're asking me to hurt someone's feelings. I mean, you could say, like, don't be Scott McBrien and that would be fine. What do they do? Write Star Wars fanfiction? I don't know. I'm kidding, of course. I love your Star Wars fanfiction. What do they do with REL? I mean, like install packages from Apple and expect it to be supported. I mean, that, you know, stuff happens. Well, I mean, like, so Apple is amazing. Oh, it really is. I'm not trying to bag on Apple. And I love Apple. And I don't know, use REL and not tell us what they don't like or do like. One of my, whenever I get, I get called into, you know, customer calls. And sometimes it's just a, you know, it's like, it's part of a various different kinds of engagements. Sometimes it's, you know, this customer is angry and they want somebody to talk to. Or sometimes it's, hey, can you explain why a customer should use this instead of this? But one of my favorite things, and I'm very open with every customer that I've ever spoken to, is getting the bad feedback. So I guess one of the things that I don't like is when people use something, they're not happy with it and they don't tell me why. Because, you know, in fact, earlier today, I was on a phone call with a customer and I was very explicit about saying that I want to know every problem that you have. Because if I don't know, I can't fix it. And it's my goal to make REL Workstation better all the time. I want your users to ask for REL Workstation because that tells me I've done my job right. But I can't do it if people don't tell me what's wrong. So the thing that they do with REL all the time that you wish they wouldn't is silently suffer. Yes. Yes. Not speak up. Yeah. And, you know, it's, some people are really not shy about giving me bad news all the time, which is, thank you, sir. Yeah. But it's, and that's great, but the danger there is that I, you know, I have to get other end users and organizations input. Otherwise, I wind up developing, you know, Red Hat Enterprise Limit Workstation for Scott McBride. And that's a really long title and it wraps around the side of the box. So that's not exciting. I would much rather, you know, make sure that it is suitable for lots of different things because, you know, I love seeing the kinds of, I love seeing the things that everybody makes with REL Workstation. It's amazing. So I want it to work for everybody. Sounds good. Awesome. So Chris, I saw that there's some, like, chatter in the, in the chat, anything we should more largely engage with, or should we pivot into the demo? Fabio, I really do appreciate your questions, but I just don't know how to answer them. Right? Like, so, I mean, I could read it on air if you want, Scott. Yeah, I can't see them. So. Okay. Right. Like, that's what I mean. Like, let's get Bob's answer on this if he wants to. The point of BSD users argument is that they have a different development model than Linux. And I realize that the BSD argument is true. What do you think about this from the point of information security? Me not knowing any BSD developers, I really don't know the answer to that. And I feel like our information security practices are at least on par with the BSD communities. If not better, hopefully. I don't know. I mean, they approach a variety of technical challenges differently. I don't know that I'd feel comfortable saying we're more secure or less secure than someone else because we just approach it in different ways. It's the same outcome, I feel like, right? Kind of get to the same point, but different ways of doing it. So yeah, let's continue forward. All right. So I wanted to pivot. Actually, let me share a screen here to tell folks that we're doing something a little bit different with our Rowel Labs. So you should be able to see the Red Hat Enterprise Linux 8 lab page. We've been working on a new series of labs called Admin 101. And as we publish new ones, you'll see that they're always in this right hand column of the page. And it's geared towards topics that are more introductory to administering Linux. So if you are new, that would be a great place to jump in and see some things with how to get started with Linux. This configuring network interface is one is about the network manager, command line interface, or NMCLI. And then there's this one on file permissions that goes into what are file permissions and that manipulating move to mod. And then when you're using to mod, do you use symbolic or octal notation? What does that mean? Managing user and group accounts, so creating users, creating that shared and adding users to it. And I know that the content author on this is Jared Anderson. And he's working on one on installing software that I think is actually, was it already published? Let's see. Oh, there it is, installing software. So it's like some basics on YUM and RPM. And then there's another one that he and I were talking about today on process and service management. So a little bit on like using PS and controlling processes, but then pivoting into after you've installed a piece of software like Apache, how do you start it and run it with the system CTO utility? So if any of those are interesting, they're all right here on this right hand column of of the labs interface for REL and they all carry this admin one-to-one tag. So I just wanted to give a shout out to Jared. The other thing that we've done recently down at the very, very bottom is I've created an open lab, which is essentially it's like, hey, here's a box and you can do things. So how long does it last? Because we have the same, we have a similar thing in the OpenShift playground. We call it and it's like 60 minutes. It's not very long. And I think ours is 30 minutes. It's between 30 and 60 minutes. Yeah. But like if you're just trying to do something simple, I think it's perfectly adequate. You just want to install a package or you want to test what a setup of something would look like kind of thing. Yeah. Suck down a container and deploy it with Podman or yeah. So there's all different kinds of things that you can do with it if you so choose. And eventually it'll get provisioned and we can see the command line. Also, it is installed with the web console, which is provided by a cockpit product. Yep. I use that almost every day. So if you hit this guy and say that you want to view ports, I think it's 1990. Yeah. Right. So you can like log in to the web console and do graphical based administration if you've not seen that before or you want to try something out before doing it on system. The web console now has a terminal built into it. So it's like meta terminal, right? Yeah, actually, you know, if you if you want to lock down your entire network so that nobody can access anything over SSH, you could still have 9090 open and still have terminals everywhere. Yeah. So that's a different way of accessing the terminal. I think web console gets kind of a bad rap, especially amongst experienced users because they think of it as graphical and graphical is for noobs, right? And in reality, there's a lot of things that web console offers that makes it a little bit easier. Makes your life better even if you're experienced. Like for example, one of my coworkers uses the storage plugin all the time for managing his logical volumes. And he's like, it's not a matter of capability. Like he could run VGs and look at the line groups and do an LV extend and mistype the size and have to go back and fix it again and grow the file system. And you know, there's like this whole procedure for doing it. He's capable of doing that, done it for many years, but a web console is sort of like click, grow, slider to the size you want, click a button. Which to be honest with you, as somebody that rarely resizes a volume, I have to go look at the commands every time, right? Like I know there's a series and I know how to do that. I've been certified in doing it, but I don't remember all the time. If I can go and just slide something as opposed to running 20 commands or 10 commands or five commands even, that's way easier for me, I feel like. And yeah, saves me more time, which is what I'm all about. Fair. And, and like we've been making more updates to web console since rail eight was released. So maybe you looked at it real 80 or 81. You should go back and look at some of the stuff now. So for example, in the overview, we have these stored metrics. Oh, I need to install it. But it'll show you like performance metrics of the box or in the containers plugin for it, you can actually like download, run, manage running containers on the box. It's actually really slick web based interface into Pogman essentially. They've also done a lot of extension of the cockpit machines plugin. So it's much more on par with vert manager, which was our classic utility for managing virtual machines on a host on a row host. But you can do things like access the console of the machine from the machines plugin, which is pretty slick. And then also more recently under software updates. So it used to be that anytime you did an install of updates, it would just say, oh, you should remove the machine. No matter what it was, like I installed Apache update, oh, remove the machine. Yeah, no, now it'll actually be like, you should restart these services. Or if there is one like a kernel, it'll go, you should reboot the machine to get this whole effect. But yeah, so if you've not seen and you can configure auto up which I think is helpful for folks. It'd be, you know, I mean, if you've got a fleet of servers and you're like, you know, these are all dev boxes and they can just update them. You know, if there's a kernel update, I want it and I want to reboot immediately kind of thing, you can have it set up that way. It's, you know, entirely up to you. So the PC, the PCP stuff that you were looking up to install performance copilot is such a fantastic tool. But I think I feel like not enough people really know what it can do for them. Well, it captures a ton of system data. And I know that we've been a bit bumpy in our Grafana integration with it. But again, with eight four, I think we've closed that gap even further and made it easier to build Grafana dashboards to be based off of the performance copilot data. So, so if you looked at it, you know, two years ago, you're like, wow, wow, which was easier. Well, you should look at it again, because might could be easier. Yeah, Grafana provides such a good way to visualize all that information, being the guy who's all about the graphical feedback loops. Well, and quite frankly, like when you're looking at scrolling textual data system performance, like how do you actually suss out that spike or that period of time where things were not great? When you're just looking at a sea of data, right? In fact, that's what well, that's what a lot of the guys using workstation are doing, right? They're taking seas of data and crunching it down into visualizations that they can use on their workstation to make analytical decisions. That's right. A long time ago, I updated Python 3. Yeah, I like the code. So I have a Fedora server, which also has cockpit installed as it's, you know, kind of upstream of Red Hat. And it has multiple nicks. And I have those nicks assigned to do different things. So I can actually like look at utilization to kind of get an idea of, are you doing the thing right or not? Are you load balancing services correctly? That kind of thing. It's very nice to be able to see kind of traffic utilization live, you know, in and out, as opposed to up arrow, enter, up arrow, enter, up arrow, enter kind of thing. Or, you know, there's other tools for watching network traffic, but it's, for me, it's just, as part of the OS, I'm going to use it, you know. So you're looking at like network IO or disk IO? Right, exactly. Network IO, especially in and out. Because, you know, I am on a quote unlimited connection, but sometimes that unlimited connection seems to favor some services more than others. So it's just interesting to see what happens when I'm like. They said it was unlimited. They never said they wouldn't quality of service all your traffic. Right, yeah. It's just one of those things where it's like, are you able to download these, you know, new updates in a speedy fashion? What's going on here? And then, you know, test and find out, oh, some reason I'm being rate limited. Great. And sometimes I found that that's actually the receiving side of your connection, and it's as opposed to your ISP side. So this is the dialogue I was talking about, that used to be you can either reboot or ignore. And now you can actually restart those services that will need to be restarted in order to pick up the most recent updates. So, and that one included a web console update, which is why I am now being told to reconnect. Nice. Oh, yeah. So I've seen it before, haven't looked at it in a while, you should check it out. Yep. Highly right. And as Chris mentioned, terminals right there. Yeah. Account management too, right? Like, if I need to add a friend, you know. Yeah, it's here under accounts. Yeah. It's easier to kind of, I know that's the one thing that I feel like Active Directory does kind of get right, you know, showing you, hey, this user is part of all these groups and that kind of thing, right? You click on a user, you get all the data right there. You can add, change, move, that kind of deal. It's not as cut and dry like ID username. Yes, I know that's very, very specific, but unless I'm looking at like the password file, I don't know all the users. They have AD for Unix services. And essentially what they do is provide the normal Unix-y things like username, UID groups in a format that Unix boxes can consume. Red Hat has Identity Manager, which is based off of a free IPA up to free project. And it can do Active Directory integration or it can manage user accounts and there's like a really sophisticated web UI for working with it if you've not seen it before. But so that was the technology portion, although it wasn't a super in-depth technology. But that's okay. Yeah. You know, it's a holiday. Tips and tricks. Holidays coming up. Yeah. Keep it light, keep it fun. All right. Bob, any parting words of wisdom before we close out the show? Throw one on me. I like putting them all in the spot. Thanks, Scott. Make sure it's Peppy and interesting. Peppy and interesting. So yeah, going back, I really, I'm just stuck on that last question that you asked me, what do people do with RHEL? And I wish they wouldn't. But seriously, if you are using RHEL Workstation or RHEL Period in your work or in any capacity, product managers want to know what you don't like. You will not hurt our feelings. Right. We work with people like Scott. We're like made of callus now. So please, please, please, we want to know what you would love to have us improve, you know, even if it seems like crazy, you know, like, you know, what would be great is if you replaced some package, some critical, like, let's say you wanted us to replace Nome with a different desktop environment. You know what will make sure that never happens? Keeping your mouth shut. Right. And that's true for anything. You know, Red Hat is deeply engaged in a lot of upstream environments or upstream projects. So if there's a major change that needs to happen, and it fits along with that particular use and a particular project, you know, and you tell us, we can, you know, we can consider it, we can talk about what that would look like, and we can think about what we can do to make changes that would make RHEL and make working with RHEL better. That's why we're here. So please tell us all the things that you love as well and the things that you don't love so much. So good advice. I'll take that. A plus for effort. Thank you. Thank you so much for joining us today. Yeah, we really appreciate it. Thank you, audience, for your questions. Thank you for inviting me. I miss having these kinds of conversations face to face at like, at summit. Nothing feels better than getting on an airplane with a voice that's just trashed at the end of four days of talking to thousands of people. It's a different kind of feel. All right, awesome. Thank you all for joining us today. Catch us in what, two weeks, Scott? Yeah, I think so. Yeah. And we'll be back with more of your RHEL content for your RHEL ears. So sign off for now. Thank you. Stay safe out there. Bye, everybody. Thanks, everybody.