Latest / Nutanix Weekly / Nutanix Weekly: Nutanix Lifecycle Management 2.6 is Here! Check out the Top 5 Coolest Features of LCM

Transcript

Nutanix Weekly: Nutanix Lifecycle Management 2.6 is Here! Check out the Top 5 Coolest Features of LCM · Nutanix Weekly

WEBVTT 1 00:00:04.150 --> 00:00:11.980 Philip Sellers: Hello, and welcome to Nutanix Weekly episode number 71. And Phil Sellers, I'm your host for today, and 2 00:00:12.030 --> 00:00:30.899 Philip Sellers: I'm a solutions architect here at XenTegra on behalf of the XenTegra team. I want to say thanks for joining us, and thanks for spending a little time to learn something deeper around the Nutanix stack. I'm joined today with Jirah Cox.  Jirah, how are you doing today? Hey, Phil? Thanks for coming on, man. It's just the 2 of us. 3 00:00:30.940 --> 00:00:39.430 Philip Sellers: Yeah, it is, we were. we were both humming a little bit of will. Smith, just before the podcast started. our our normal 4 00:00:39.570 --> 00:00:44.470 Philip Sellers: co-hosts and compadres here on the podcast or all 5 00:00:44.550 --> 00:01:10.880 Philip Sellers: doing other things. Today, I I know we've got a variety of different customer meetings and other conflicts on the calendar. So you've got Gyra and myself for the for the next hours, so only the only the truly dedicated your head went to Will Smith. I was trying to think we're similar size, guys, you and I, I'm sure I think which of us is Dr. Evil, which one's many. Me? so yeah, we definitely could go to the Austin powers as well. 6 00:01:10.880 --> 00:01:28.720 Philip Sellers: Well, it's been a a minute since we've had a chance to get together. we talked a few weeks ago about Geo. Distribution for the object stores. And that was a really cool push forward. 7 00:01:28.790 --> 00:01:32.340 Philip Sellers: yeah, today. we, we've got one that's 8 00:01:32.580 --> 00:01:40.569 Philip Sellers: going to be in the crosshairs of every single Newtonics customer. I'm kind of excited to talk about this because it affects everybody listening 9 00:01:41.430 --> 00:01:57.270 Jirah Cox: absolutely. Yeah. So this comes to us today from the neutronics dev blog over on our community site. where we can get some of our in some ways deepest more detailed blog posts. So yeah. Glad to be able to share it today and summarize it. 10 00:01:57.290 --> 00:02:16.749 Philip Sellers: Yeah, absolutely. So. we're looking at Nutanix's lifecycle management. So LCM, it, it's one of the things that mechanics does extraordinarily well within the platform. And today the the blog post is check out the top. 5 coolest features in Lcm, so you know it. It's it's certainly 11 00:02:16.910 --> 00:02:41.790 Philip Sellers: like I said one of the things that Nutanix does really well with the platform, and I I love the way that it's been approached. you know, not only in consideration of the Nutanix software, but the partner ecosystem around it. And I think that's one of the things that really makes Lcm great. what stands out to you is kind of the the selling feature for Lcm. 12 00:02:42.390 --> 00:02:48.650 Jirah Cox:  let's see, I'm gonna go with like sort of its reasons for existence, right like it has to. 13 00:02:48.740 --> 00:02:57.369 Jirah Cox: It has to exist any system that any system that does advance things for customers like a like a cloud platform would 14 00:02:57.400 --> 00:03:09.319 Jirah Cox: has to involve complexity. Right? If if there were no complexity in system it would be easy, and everyone already have it, and there'd be a hundred different competing versions of Newtonics, 15 00:03:09.360 --> 00:03:14.070 Jirah Cox: and and of cloud platforms. So there's complexity in the system. 16 00:03:14.340 --> 00:03:36.989 Jirah Cox: But new tennis is built, and in such a way that we don't ask customers to bear the Comp. The cost of that complexity. Right? So we never throw. And we can all think back to previous lives where we've had to. Do. You know, care about Hypervisor versions and firmware versions and adapters and bios updates and storage arrays and connectivity between all of that right, whether it's Ethernet or Fc. Or whatever you want to do. 17 00:03:37.110 --> 00:03:42.089 Jirah Cox: and order operations and compatibility matrices and all that good stuff. 18 00:03:42.370 --> 00:03:59.720 Jirah Cox: good stuff, question, mark and we simply just never ask a customer to bear the cost of wrangling. That complexity right? The put complexity exists. No one can say it doesn't, but we do our very best to abstract that from the customer. It's the the customer from it. 19 00:03:59.730 --> 00:04:05.040 Jirah Cox: and let the system be as self-driving as possible. Right to sort of mix a bunch of analogies together. 20 00:04:05.160 --> 00:04:23.520 Philip Sellers: Yeah, I mean, I I think back when DC came around right? You know, this is this is a concept that they really kind of brought to the marketplace. And it was around managing the total solution, not the pieces and parks. And I remember being a practitioner. 21 00:04:23.720 --> 00:04:34.780 Philip Sellers: you know, in those days, and having to go out to websites and look at recipes and look at what? What versions of drivers worked with what versions of a firmware. And 22 00:04:34.800 --> 00:04:40.960 Philip Sellers: you had to worry about a lot of things, and the complexities only gotten deeper. 23 00:04:41.080 --> 00:04:55.689 Philip Sellers: since you know that that was a decade or more ago. The complexity is only increased. So I I agree with you when you say it's a necessary piece of the puzzle. but not all of your 24 00:04:55.710 --> 00:05:00.700 Philip Sellers: Competition makes it as easy. So the simplification factor is huge. 25 00:05:00.810 --> 00:05:27.280 Jirah Cox: sure. And we and we you know, we played our strengths right? We do as much as possible as we ever can in software, right? That's kind of our mo for designing anything. so whether that's you know, storage or virtual networking, or, of course, the Hypervisor and and Storage Controller interaction. You know the patching itself as well. you know, software eating the world right by by byte. so we get to. We get to. You know. 26 00:05:27.370 --> 00:05:37.949 Jirah Cox: we get those tailwinds right in our favor of yeah, we can do the lowest levels. Right? Nick, firmware, Bios, firmware CPU micro code. All that good stuff that 27 00:05:37.960 --> 00:05:56.330 Jirah Cox: we'll kind of never go away because software has to run on good hardware and the rest of it is an availability model that to to the largest degree we write and control and can articulate, and can orchestrate. Sorry what I meant to say. in software, right? So software defined and failed software to find availability. 28 00:05:56.330 --> 00:06:12.239 Jirah Cox: let's us make these upgrades, you know, easy for our customers, because they just hit the go button and and deliver for their workloads right by doing storage availability or live via migration, or whatever the update requires. all automatically for them. 29 00:06:12.440 --> 00:06:18.379 Philip Sellers: Yeah, that orchestration is a huge piece of it. It's a time saver. It's it's also 30 00:06:18.560 --> 00:06:47.900 Philip Sellers: an autopilot feature, right? I mean, at the end of the day it's a lot less you have to think about and coordinate all of the bringing months, and new tenics have already thought that process out for you. Yeah, I mean it. It's We've always said, you know. I've I've always said our our true North. For, like directional engineering, like what we want to give our customers is that cloudcom, but anywhere right in a cloud or outside of cloud, or, you know, helping your collection of racks in a data center become an act like a cloud. 31 00:06:47.900 --> 00:07:00.569 Philip Sellers: and to the large degree possible that involves stuff like, how much can we help you stop worrying about when it comes to? Oh, I gotta go apply the updates. No, you you kind of. Don't you gotta go tell us to start, and we'll we'll handle the rest. 32 00:07:00.570 --> 00:07:19.170 Philip Sellers: Well, that's a great segue directly into the article. so I I kind of want, sound effects, or something like that, you know. It's like, Hey, you know the blog post. We're thrilled to announce a new version of Newton's lifecycle management version 2.6. So today Jyra and I are going to go through those 33 00:07:19.170 --> 00:07:33.720 Jirah Cox: and and the top 5 features. So why don't you kick off with the first one. Yeah. So with gratitude to the post author, right? Micro Monday, who actually trivia, Mike and I went through new hire together at Newtonix, and it'll be 34 00:07:33.980 --> 00:07:44.909 Jirah Cox: 6 years ago in a couple of months here. So Mike's a good friend, he's a great guy, so I love that he wrote this blog post for us. He probably didn't know he was doing it. Just so we could put it on a, podcast but it turns out he was 35 00:07:44.920 --> 00:08:07.819 Jirah Cox: first bullet point here. Right? So So now Lcm, powered by pre upgraded upgrade. Pre upgrade and update checks, so you can do even more stuff like a dry run. or more, like a pre-flight sanity check to say, when I go into a place that apply these up upgrades, or what's in my selected like cookbook of what I want to apply on my next patching run. 36 00:08:08.360 --> 00:08:33.210 Jirah Cox: Then, Lcm, the lifecycle module bicycle management module in, you know, Npc. Or in prison element, can do those upgrade checks to ensure that we're giving our customers like a higher degree of success, right? That can include stuff like download the binaries. Now, right? Maybe you know, let's safeguard you against like an Internet outage during your maintenance window. Let's get those binaries now validate against those kind of things. The article talks about. 37 00:08:33.750 --> 00:08:49.960 Jirah Cox: those modules can even run Ncc pre-checks. So Ncc. More. 3 letter acronyms to throw out the audience is the Newtonics cluster check. It's like a something like 700 different health checks that every cluster runs on itself autonomously. 38 00:08:50.040 --> 00:08:51.980 Jirah Cox: The checks are written by our support team. 39 00:08:52.020 --> 00:09:04.459 Jirah Cox: upgraded or released on about, and every 6 week cadence about every 6 weeks there's an there can be a new health check script if we found anything that your cluster should learn how to check itself against preventatively. 40 00:09:04.550 --> 00:09:11.880 Jirah Cox: It's almost like having one of our Rsrs or site reliability source reliability engineers embedded in your cluster right? Looking for stuff proactively 41 00:09:12.460 --> 00:09:14.570 running like 700 different checks 42 00:09:14.660 --> 00:09:21.440 Jirah Cox: on various intervals. So some run like once a minute, some run like once a night, depending on how often could this happen? 43 00:09:21.630 --> 00:09:35.539 Jirah Cox: but So this now pre upgrade can even run those health check scripts catching anything that could throw a wrench in your patch cycle like, say, once you get into your maintenance window when time is much, much, much more precious. 44 00:09:36.390 --> 00:10:00.820 Philip Sellers: Well, not only just precious, but, you know, from a practitioner's point of view, you may not be firing on all cylinders, because you're up at 2 in the morning without coffee, without caffeine maintenance, too. So better to do that during normal business hours, and know there's an issue. So you can remediate before you get it. That maintenance window. Yeah. Iron out those circles now. Absolutely So 45 00:10:00.930 --> 00:10:07.060 Philip Sellers: the article talks a little bit about this. What software is available And supported for the pre upgrade checks. 46 00:10:07.840 --> 00:10:30.429 Jirah Cox: hold on, This is the sound of gyro getting to where you're looking on the page. the so they can do this for yeah. So for Aos, which is our the hey? This is what it runs within the Cbms. Right? That's the storage OS within the within the cluster itself. Prison central. The multi cluster manager, and can also do. Dry runs against Esxi as well. 47 00:10:30.560 --> 00:10:35.719 Philip Sellers: That's pretty awesome. So not just the new tank suite also. also. 48 00:10:35.900 --> 00:10:50.519 Philip Sellers: Esxx, yeah, more to come on that one in just a minute. Absolutely So the next one up foundation joins the party. You can now upgrade new tax foundation in parallel to other upgrades with Lcm. 49 00:10:50.800 --> 00:10:54.440 Jirah Cox: so foundation is sort of the collection of code that 50 00:10:55.060 --> 00:11:22.040 Jirah Cox: takes like, say, bare metal at nodes, nothing installed and makes them into genetics, nodes right? So that that runs on every node by definition at least once. It's once during its life cycle, maybe multiple times. whether you're running nodes that maybe come from the factory with no, nothing installed like I don't know. Say, a Cisco like Ucs which is still validated in a if you buy this exact acl. You can run the tanx on it kind of way. 51 00:11:22.110 --> 00:11:49.839 Jirah Cox: or even more things that are more tightly integrated like an oem offering like HP. If you want to convert them from the factory running Hv. To into Running Esxi. Then you're in foundation to say I need to change out the bare metal, hypervisor installation there, and then make it into an X node again, right running a Cvm with the proper storage plumbing, and all that good stuff so foundation can run on the node at least once in its lifetime, maybe multiple times in its lifetime. It's that sort of imaging workbench. 52 00:11:50.260 --> 00:12:14.019 Jirah Cox: And that's also where, like, let's say, you ran a super, old cluster. It's like 5 year old hardware. you. But you've you've stayed up to date on your software upgrades because we make it easy for you to do that. So you're running late in the latest creative software on the old hardware. And then you order your replacement nodes right? You call up your you know your friendly rep and say, Hey, I want a new hardware for this cluster. That's still doing a great job for me. 53 00:12:14.020 --> 00:12:30.149 Jirah Cox: I want to eject the old hardware when that new hardware lands in the rack. It's the foundation bits that know how to teach the new nodes how to get on the least and greatest software and firmware, because that's newer nodes and then join the cluster. So teaching even the old cluster. How to run the latest and greatest foundation is why you kind of want that. 54 00:12:30.440 --> 00:12:46.760 Jirah Cox: And so as part of our multi year evolution for Lcm, it's changing in in a lot of ways. Right? It's it's replacing functions we've had in the product for a long, long time. Right? 6, 8, 10 years. In some ways we've sort of always made easy upgrades and emphasis for our customers. 55 00:12:46.810 --> 00:12:58.379 Jirah Cox: centralizing a lot of that into Lcm. And making the upgrade process even smarter, and putting even more crowd rails into place. It's kind of a hallmark of when things come into Lcm, so that we can now add foundation 56 00:12:58.670 --> 00:13:02.730 Jirah Cox: to to upgrades that Lcm can now control for customers. 57 00:13:03.640 --> 00:13:31.470 Jirah Cox: Yeah, another change. It looks like you. You're also taking into account your dark sites. So not just your Internet connected side and keeping them updated. Now, Lcm, is a bundle portion above cranes for your dark sex. Yeah. So in the earliest iterations of Lcm. If it's been a minute for for anybody listening since they first touched Lcm. When I'm first launched it was sort of the the first iteration kind of assumed Internet connectivity. 58 00:13:31.510 --> 00:13:45.279 Jirah Cox: along the way. It's not necessarily new in this release. But this payload probably is it did get the ability to do dark side stuff where I can. You can either upload your own patches directly. So you look at the benefit of the Lcm patch application engine 59 00:13:45.310 --> 00:13:56.070 Jirah Cox: and an orchestration and order operations. Awareness. even though you're feeding. It Updates manually as a human right, bringing updates over on a thumb drive, or whatever your air gap needs to needs to look like. 60 00:13:56.070 --> 00:14:19.219 Jirah Cox: or you can host your own internal repo. Right? These are my blah blah blah and to so ink in approved updates that are approved for internal release. I'll host them on a website and any cluster in behind my firewall can then fetch from that repo source and do that as well. So we consider both of those a dark site, either either a customer managed internal repo, or or literally air gapped. And you're just hand-feeding it updates. 61 00:14:19.510 --> 00:14:36.490 Philip Sellers: that's awesome. you know, you're you're not ignoring Esxi customers, either. there's a some security features that are are introduced in 2, 6 as well for your vm, our customers? 62 00:14:36.910 --> 00:15:01.539 Jirah Cox: Yeah. So the the cool thing there is. We can now. yeah, I thought, this is new in 2,006 as well, like even just flat out the hallmark release there. Lcm, can now do Esxi. Right. That's that's cool, and it of itself a much, much easier workflow for our customers. Then, again, the old school workflow from the that. This replaces our our old school. One click hypervisor upgrade. 63 00:15:01.540 --> 00:15:16.510 Jirah Cox: as well as, of course, way easier than using like native view or patching, but telling it oh, wait no pause, let me go. Put this Cbm. And maintenance mode, and then do the whole T ceremony, and you become the orchestrator? Right? It's not a great place to live as a human being. so now, with that ability to upgrade. 64 00:15:16.510 --> 00:15:36.650 Jirah Cox: I right through Lcm, and then, of course, do it securely as well. Right? So it talks about uploading image check sums as well with those Esxi payloads to ensure that you've got a healthy supply chain for your software. which, of course, is what we do for Hv. as well. That lets you get the benefit of those easy upgrades without compromising on security, either. 65 00:15:37.230 --> 00:15:40.580 Philip Sellers: That's a huge step forward. yeah, I mean. 66 00:15:40.830 --> 00:15:42.760 Philip Sellers: the Vmware 67 00:15:42.800 --> 00:15:59.029 Philip Sellers: delivers nice upgrades, but they don't necessarily take into account the Cvm. And and things like that. So you're right. You end up as that orchestrator moving things around within the cluster. If if you're a Vm, our customer today. So this is a huge 68 00:15:59.120 --> 00:16:02.599 Philip Sellers: huge thing for any other. Vm, our customers listening 69 00:16:03.850 --> 00:16:26.490 Philip Sellers: the inventory checks. you know, this is one of the great things that I like about Lcm is is, you can always go in and run an inventory, and it's gonna give you details of of what you need, not only from a software stack, but also the firmware. And you know any oem drivers things like that. So 70 00:16:26.740 --> 00:16:36.590 Philip Sellers: number 4 on the list of our top 5 is inventory checks at Lunar. Chris speeds. I love that. Yeah, I mean, 71 00:16:36.860 --> 00:17:06.039 Jirah Cox: I couldn't come in. If that's trademarked or not, it's certainly fast right? The well, of course, is ludicrous speed. A trademark from a car company in the 2,000 twenty's, or from a sci-fi moving in the 19 eighties. Right? the either way. Lc, I'm running fast is a great thing for customers. because inventory is, let's call it. it's thorough, right? So it's going to show you that report of like, here's all your nodes for large customers, right? Large clusters might take even longer to run. But like here's every 72 00:17:06.040 --> 00:17:24.239 Jirah Cox: bit of firmware on the nicks, and on the disks, and on the board, and on the out of band. And all this great stuff. It's it's very thorough detail could take a minute to run right, or it could take can at least several minutes to run. Some customers even automated to just run on there on its own at 3 in the morning every morning, so that when they come in the office 73 00:17:24.349 --> 00:17:52.070 Jirah Cox: They can see the latest and greatest detail there without a need, for you know, waiting on that as a human to go run that that's probably let's call it an ideal scenario. But of course there are times when you, as a human, want to go run that or or the box wasn't checked to auto run every morning. So then you are waiting for it. So it's so good for it to run faster. For those times where you are waiting on it because you're gonna act on the results and tell it to patch like right right away. So so yeah, certainly good to always be 74 00:17:52.070 --> 00:18:21.590 Jirah Cox: making that leaner and meaner from a execution time just eyeballing the chart here that is embedded in the blog post. I'd say it looks like it's about. I don't know 30, 40% reduction. It's gonna give me 50. I was, gonna say, you're approaching 50% better. I mean, it, it looks like half to me. I mean, it's a little bit more. Oh, this is a tech. Podcast. Phil, if not marketing, you're lucky to get out the ruler. Right? Exactly right. We we need more granularity in the chart to be able to. 75 00:18:21.590 --> 00:18:40.729 Philip Sellers: That's right. But that's yeah. I mean to your point. It's a great thing to have this moving faster. you know, as I go into health checks with customers, this is one of the place. We always look one to make sure Lcm's calling home if they're connected, and make sure that they're getting updates, but also to see you know how far out of 76 00:18:40.730 --> 00:18:56.980 Philip Sellers: current, are we, you know? Are we on a short term release? Are we on a long term release? There's a lot of different things that Lcm helps me do a as an advisor to our customers, too, for sure. I mean. 77 00:18:57.050 --> 00:19:03.410 Jirah Cox: you know, my my giving PC. Next to us here is supposed to update itself in the morning at at night, when I'm asleep. 78 00:19:03.790 --> 00:19:14.089 Jirah Cox: That doesn't always happen. And so when it does happen to have to patch mid day. I want that to be as fast as possible. Yeah. yeah, nothing worse than watching the little spinning. Please wait 79 00:19:14.120 --> 00:19:26.460 Philip Sellers: and number 5 on our list. Lcm, final mentions so small, very short section here. But I'll give it to you. 80 00:19:27.430 --> 00:19:45.730 Jirah Cox: Yeah, so so so faster. Self auto upgrades, which is a weird phrase, to use what that means is so? And the way Lcm operates right when you go to tell Lcm. Hey, go scan the cluster to run my inventory, tell me what I could, what I have installed, and what I could upgrade to 81 00:19:45.940 --> 00:19:59.229 Jirah Cox: the very first thing it actually does. Every single time is upgrade itself right? Because we want to give our customers the outcome of always applying upgrades, even if you're moving from like N minus 3 to N minus 2 firmware. If you're like. 82 00:19:59.340 --> 00:20:04.970 Jirah Cox: super super change, reverse risk averse. You want very, very old 83 00:20:05.000 --> 00:20:14.530 Jirah Cox: baked stable things. You can do that with the latest and greatest upgrade logic, right? So Lcm upgrades itself first to the latest version. 84 00:20:14.580 --> 00:20:24.240 Jirah Cox: so that it catches any new logic or an operations. Compatibility matrices, bug fixes anything that entering wants to put into it. 85 00:20:24.240 --> 00:20:45.360 Jirah Cox: you can apply even those like, say Ms. 3 to Ms. To upgrade an old version of code with the newest logic in there. And so lc, I'm upgrading itself faster. as well as giving you more visibility into the process again. Good for our customers right? So that if something does go bump or something and a ring list to get ironed out, we want to give you the actual details right there, up front. 86 00:20:45.360 --> 00:20:48.850 Jirah Cox: or if everything goes smoothly and it's all boring, then make it as fast as possible. 87 00:20:49.130 --> 00:21:00.439 Philip Sellers: Yeah. And I I love this upgrading the upgrader. That's the way Michael puts it in here in the blog post. So love that I mean you know it. It it sums it up perfectly, I mean. And 88 00:21:00.450 --> 00:21:29.070 Philip Sellers: again, I love what you're saying about. Why that happens. The logic upgrades the the rationale, you know, across the board as Newtonics as a company learns things. They're trying to put that back into the automation and and into the customer environment, so that more customers benefit from the knowledge of of whatever's found within the software or the support experience. So that's fantastic. 89 00:21:29.340 --> 00:21:39.680 Philip Sellers: you know. The the final words on the article really talks about our battle against cyber threats. And this is this is a huge thing for us, I mean. 90 00:21:39.810 --> 00:21:55.070 Philip Sellers: you know, I we at Z Integral will will tell folks it's not a matter of If you get ransom where it's probably a matter of when will you be affected by ransomware? And how are you preparing for that? How are you? 91 00:21:55.120 --> 00:22:08.889 Philip Sellers: defending against that? And and one of the best things we can do as practitioners out there running systems is to stay regularly patched and updated with our system. So, 92 00:22:09.140 --> 00:22:30.690 Philip Sellers: you guys that Newtonics are are helping deliver an easier experience. more automation and just reducing friction overall and making that happen so. great, great stuff for us. What? I'll I'll turn it back over you for final words. To what? What are Jyra's final words for this. 93 00:22:30.730 --> 00:22:49.340 Jirah Cox: I I think about The Platonic ideal of updates is probably your cell phone like, it's hard to imagine the last time that that got wedged or went sideways or needed any kind of support. You just you see, the button updates are available. You hit it at the right time for you cell phones aren't highly available. So it's even 94 00:22:49.440 --> 00:23:17.210 Jirah Cox: more of a painful process as the cell phone owner compared to Linux platform owner, right where this is like not impacting your running workloads. But that's I think that's the right kind of experience we're gonna give. Our customers is like safe trusted updates that, you know, do good things for you from either a product capabilities or security standpoint. And and hopefully, you know, we're doing our best to to tip away at well earned decades of industry. Mistrust? 95 00:23:17.290 --> 00:23:38.959 Jirah Cox: Oh, no, not updates. Again. Let me put that off as long as I can. Your point. We we have to help do our part to help get customers in a mindset of updates are good, updates are easy, and they're good for me. and I should do them, I should stay on top of them. when the manufacturer advertises them right? That's then we do the same thing that even smartphones do around. 96 00:23:39.010 --> 00:23:44.769 Jirah Cox: you know, like, when you hear about the new Ios blah blah blah version is available, I could go fetch that update on day. One 97 00:23:45.080 --> 00:24:10.010 Jirah Cox: apple will push that to me probably 3 weeks later. Ish right when they're like we're past the Cdn slam launch or past the day 0. let's just say findings right? I probably want you know I don't. I/O 16. I want 16 dot one or dot one or something like that. they have it in their control to say, when do I go poke a user and offer the update versus? What is it available to fetch? We have the same model. Right? We put updates on our on our portal. 98 00:24:10.010 --> 00:24:23.340 Jirah Cox: on day. One is when we consider a lot a release to be Ga, and then we separately. Have a have a release mechanism where we can say, now, we're going to push to your cluster, and it's considered one click installable and advertise that is available to your cluster. 99 00:24:23.370 --> 00:24:44.989 Jirah Cox: based off of it. Communicated support controls that experience. Right? So they're at the at the nerve center of everything right from what are they seeing? across, you know, performance across support cases, across user experience, everything. And they they have it in their control to give it the blessing to say that's now released out to the field for broad consumption. And it we were we were. We were the same way trying to go for that same sort of 100 00:24:44.990 --> 00:24:55.539 Jirah Cox: availability model and making updates. A let's just say as pleasant as possible in experience. They never get to be like yay, I did them but they can get, you know, pretty pain free. 101 00:24:55.870 --> 00:25:23.839 Philip Sellers: Well. And and again, I mean, you know, I I your cell phone examples actually really great. Because, you know, I I think about OS upgrades, and there's new features. There's new things to consume. And so there's a desire to get to those sooner. And so you're right. Apple. On day one, their Cdn's pretty saturated. It's pretty slow to to get the updates, because there's so much demand for it 102 00:25:24.150 --> 00:25:45.230 Philip Sellers: as opposed to living through the days where it's like, I'm not installing natural service back one comes out, you know. you know we we all have been burned by different software patches and upgrades. And you know, we we're we're in a more calculated approach, I guess, is is the best way to connotate it 103 00:25:45.230 --> 00:25:56.109 Philip Sellers: with updates and things, so that we're able to to get those features out in the hands of customers in a safe way. But at the same time, you know. 104 00:25:56.370 --> 00:26:00.519 Philip Sellers: learn from it, and then take a cat, you know I'm more 105 00:26:00.790 --> 00:26:22.580 Philip Sellers: intentional approach to to getting it into the hands of customers like you said, I love the fact that support is is a factor in that. If if they're seeing, you know not great things about a particular release. They're able to affect change. They're able to, you know. Give that feedback into the loop. So 106 00:26:22.650 --> 00:26:25.510 Philip Sellers: all of that works together for the customers benefit. 107 00:26:26.850 --> 00:26:27.520 Jirah Cox: Yeah. 108 00:26:28.780 --> 00:26:52.870 Philip Sellers: Well, Jarra, as always, I've enjoyed catching up with you for everyone listening. And I hope that you've learned a thing or 2, and like. I said, we had a topic today that affects everybody. We all need to patch. We all need to upgrade, and it's tanx has given us some great tools and some great abilities in this new version of Lcm. So, Gary, I want to say thanks on behalf of Zen Tiger, as always, for joining Us. 109 00:26:53.050 --> 00:27:11.200 Philip Sellers: Thanks for having me man. And thanks to Mike for this blog post. Absolutely. Thank you. Mike, shout out to you from the new tanks, weekly podcast crew. We do appreciate everyone that gives us this great content that we talk about every week. But mike, this week's for you. So we do appreciate it. 110 00:27:12.060 --> 00:27:17.540 Philip Sellers: I guess. I guess that's a closing for us. I I want more time with Jara. 111 00:27:17.960 --> 00:27:25.209 Philip Sellers: Well, if you're behind on your podcast. Then you might have a new episode in a few seconds. Here we'll see 112 00:27:25.260 --> 00:27:33.980 Philip Sellers: all right, everyone for Zintagra and the new Tanks team. I want to say thanks for joining us, and we will see you on the next episode.