DOP 290: KubeCon North America 2024 Review
38:40 Share

I think that now we are more, of course, there are new projects, so I'm not saying that it's dead, but it's more about figuring out, okay, but what are the choices? And what are the best practices and what are the frameworks? And there is a pretty big explosion in platform engineering activity because it's no longer about the projects themselves now, but how to best utilize them with one another. This is DevOps Paradox, episode number 290. KubeCon North America 2024 Review.

Welcome to DevOps Paradox. This is a podcast about random stuff in which we, Darren and Victor, pretend we know what we're talking about. Most of the time, we mask our ignorance by putting the word DevOps everywhere we can and mix it with random buzzwords like Kubernetes, serverless, CICD, team productivity, islands of happiness, and other fancy expressions that make us sound like we know what we're doing.

Occasionally, we invite guests who do know something, but we do not do that often since they might make us look incompetent. The truth is out there, and there is no way we are going to find it. P.S. It's Darren reading this text and feeling embarrassed that Victor made me do it.

Here are your hosts, Darren Polk and Victor Farsen. Here we are yet again at KubeCon US or NA or whatever two letters we want to use. This is 2024's edition. Victor is here.

He can't talk, but he's here. No, no, I thought that's a statement, not something that I should respond to. Yes, I'm here. And we hear our jovial laughter of Whitney. Whitney, how are you doing? I would have a hard time hiding, I think, if there's anything funny happening around me. Can you notice how kiosk-yuk are you doing, but not me? We've all learned not to ask you how you're doing. Good, good, good.

Good, good, good. Same as ever. Yeah, that's the way it is. Always good, yeah. Let's just jump to the end real quick. How was it? And then we'll rewind a little bit. In general, compared to, let's say, against U.S. last year, which was in Chicago, I think, right? Probably just as good or better than U.S. Yeah. I don't think it matches Europe. No, no. But, yeah, compared to last year, U.S., yes, more or less the same.

Yeah, I don't think we can compare U.S. to Europe ever. They're just completely different animals, I think. Yeah. But this one in the end, I think, had about 9,500 attendees. Something like that. The numbers that came out. And I think Europe was like 14K. And I feel it being a little slower. Yeah. But still really nice.

Do you know what else makes it feel slower? The fact that we're in a ghost town. There are no people in this town outside of KubeCon people. It's very weird. Well, we'll get to where KubeCon is going next year in a few minutes, but...

It sounds like you had a good time. You're ready to go home. You have no voice left, which is standard for basically day six, day seven now at this point. Oh my God. That's a proof that we've been here. Otherwise people would doubt. If anyone says I feel well rested. Yeah. You're a liar. You're a liar or you're not experiencing KubeCon fraud really. So I've been watching from afar. I've been watching some of the releases and,

Nothing seemed completely crazy to me. Seemed pretty much in line. I saw Dapper graduated. I saw Cert Manager graduated. I saw new versions of Qvert. But nothing to me that was exciting, for lack of a better term. Am I wrong in that? I feel that this is in line with what I've been saying for a while. I think that within this space...

cloud native we're in a consolidation phase right we're not in a phase anymore where every year you go to pubicon and you hear 50 new ideas that you haven't even thought of and uh 75 new projects that are doing the same as other projects but slightly different and so on and so forth right i think that now we are more of course there are new projects so i'm not saying that it's that but it's more about figuring out okay but what are the choices

And what are the best practices and what are the frameworks? And there was a pretty big explosion in platform engineering activity because it's no longer about the projects themselves now, but how to best utilize them with one another. How much AI was talked about? Was it as bad as last year or was it a little more subdued? I can comment only on the expo floor and Whitney will have to chime in for any of the talks or keynotes.

I had a feeling of less AI and I was very surprised to be honest. I thought it would be AI, AI, AI, AI, but I feel that I haven't seen much, if any AI to be honest. I mean there are, but nothing extraordinary.

It's almost as if the companies figured out, "Okay, so we cannot fake it, really. That's not really a thing either." I got so sick and tired of hearing AI chatbot. Like the use case example for AI is almost always a chatbot. And that even happened during the keynotes. Yeah. And once you get past the really low-hanging fruit, like what are you really doing with AI and how is it making a difference? Yeah, exactly.

Doesn't matter whether we're talking about KubeCon or something else. It's what Whitney just said. Can we figure out what's AI for first? Yeah. So in the keynotes, AI was talked about a little bit, but it definitely wasn't like...

In Europe, AI was every single keynote. It actually annoyed me a lot. And this time they didn't bring it up for a while. Or as Victor pointed out to me when we were talking to each other, it was indirect. So they were talking about how to run big batch workloads. So they never said AI, but the use case for that is AI. Yeah. Yeah.

But they went a good 30 minutes of keynotes without mentioning AI. I think it was great. So here's my theory of the AI, right? At one moment, the hype will slow down or disappear. It will become a thing like anything else is a thing. And at that moment, we are going to finally figure out that it's mostly the same.

you still need to run it on servers you still need to manage it somehow you still need to have networking around it and storage a lot of storage and so on and so forth right now obviously there are some differences but differences are not as big as we think they are like hey yeah you need to select gpu instead of cpu okay but

is still more or less the same. The major challenge is probably the sheer volume of it, right? Now you're talking about something that could be measured in hundreds of workloads is now thousands, right? Or tens or hundreds of nodes are now hundreds or thousands of nodes. So I feel that the major challenge there is really the scale itself.

They announced a project or a Kubernetes subproject called Q, which is a K-U-E-U-E, which queues workloads and batch drops across clusters. So the AI was more like how to run AI when it did come up than it was what to do with the AI itself. One thing that I saw announced either today or yesterday was GKE being able to support

65,000 worker nodes. Did you all see that? No. I saw it. I think it was announced actually even before KubeCon a few days ago. But anyways, yeah, I saw it. What are the implications of that?

More nodes. Well, thank you. What can you do with more nodes? Run more stuff? Run more stuff. That's the key thing. So the headline for this was 65,000 nodes and counting. Google Kubernetes engine is ready for a trillion parameter AI models. So that's great. Yet more AI models. But if anybody's going to do it, it's going to be Google, right? They're bailiwick.

I don't understand sometimes what's a sponsored keynote and what's not. So there is something they talked about. Oh, I can explain it pretty easily. They all are. They all are. Like there was a mention of an AI hyperscaler. Like somebody is trying to be an AI hyperscaler. Whether it was a sponsored keynote that I'm gullibly putting here. But I thought that was interesting.

Because the alternative would be that the community chose AI hyperscaler as a choice for the keynote. Could the community be doing that? No. Okay. I feel, I've egged on my face. Never mind. I'm not going to mention anything about it. It's too late. Okay. Moving on. So we know that Victor didn't go to any sessions or didn't go to any keynotes. So we're relying on Whitney to tell us what actually happened.

So what did happen? Was there anything in the keynotes that sort of made you think, okay, that's cool, or here we go yet again? Before Whitney answers, what you said is not true. I went to the first 15 minutes of the first keynote. It's not that I haven't been on any. The first 15 minutes of the first keynote was really weird, actually. Do you remember it? Yeah. It was like, hello, welcome to KubeCon. We should all be afraid of patent trolls.

And even still, I was like, Pat and Charles, what are the syllables coming out of her mouth? I couldn't even figure out what the heck she was saying. And when I say she, I mean Priyanka. And then it was just like, the open source community has a problem with malicious actors

creating patents and trying to steal the intellectual property. But don't worry, Linux Foundation has a partnership with blah, blah, blah. And we have you protected. And I was like, I don't understand. You're telling me there's a problem. You're telling me there's a solution. And it's the beginning of coupon. You're opening with this. Why am I like, why? Why?

First you scare children and then you feed them. It's really important they demonstrate value hard and fast. I thought it was strange. There's this illustrated troll that they put up on the screen. It was very weird.

Actually, there was some illustration with a bunch of bulls or something behind the troll. Oh yeah, there was oxen. Yeah, but that illustration made me feel sorry for the troll. Honestly, there were menacing bulls.

They look mean. Whatever type of animal that was, they defend themselves by making a circle where their butts are in and their horns are out. Yeah, but each of those animals was bigger than the troll. Each of them individually could crush that creature. Yeah, we gotta crush this terrible bat and troll it.

Yeah, well, like, okay, we love to talk trash about the keynotes. But what in your mind would be a good keynote? Like, what should they open with? The first day should be dedicated exclusively to the projects. I agree with that. We should hear what's new, what's happening, what's going on. We should see cool demos. It should be tech-oriented and so on and so forth.

I completely understand and agree that there should be sponsored keynotes because money needs to come from somewhere. That's all good, right? Yeah. But not necessarily as the opening act. I would move those to the second day, right? Because you want the first and primary focus of the open source foundation must be open source. If that's not the focus of the foundation, why do we have the foundation, right? Yeah.

And yes, sponsor talks should be there. I understand that completely. But it should be maybe not for starter. Yeah. That's all day three now. Day three is pretty much 100% community and which projects have graduated. Cool, but kind of it makes, it leaves bad taste in my mouth kind of that, okay, at the very end, okay, by the way, this is open source.

Yeah. And if you want to highlight the value of the open source foundation, maybe not patent trolls, but instead like the awesome humans behind what's going on. So what awesome humans did you see there? Okay.

All of them. Can you think of any? All of them? Well, you know, KubeCon is a place where everybody is, right? Yeah. It's twice a year, not counting Geyser. There is a meeting point for everybody, right? From Susie, RedCat, startups, hyperscalers, you know, all the friends and colleagues and enemies and everybody else is there, right?

So yeah, the answer, everybody sounds just about right. Yeah. And when you say who, I mean, we're talking about just like open tech celebrities. Kelsey Tower, I was here. Who else even is a tech celebrity? Liz Rice is around. I will say the security celebrities, a lot of them have boycotted this. CubeCon, because it's in Utah, which doesn't have friendly laws toward trans folks.

And so they felt unsafe or unseen and decided not to come. So there were some holes. So we talked about the people. We talked about the keynotes. Any good sessions, Whitney, that you went to? Because I know for sure Victor did not go to any sessions. No.

I'm not the best at going to sessions either. Oh, you've been hanging out with him too long. I don't know that I was ever very good at it. Except that there was a keynote actually that was paving, that was relating platform engineering to generative AI, which I thought was, I didn't, I don't know that they said something super new about platform engineering, but I did think it was nice to relate those two things to each other because it's

I don't often hear of them together, and they are very interconnected. Did either of you learn anything new this year? Or maybe not necessarily new, but made you think, oh, I hadn't thought about that. I need to revisit this again. From my side, no. It's going back to my consolidation story. Well, for you, though, Victor, AWS made a pretty big announcement that it fixed. Yes.

In the cross-planet space, right? No, it's just, but it's another project that is similar to other projects, including cross-planet, right? So to me, that's not new. I mean, if new is a new project, yes, right? But yes, there is a fifth variation of that thing, right? But when you ask about new, I assume that, I thought that you're referring more like, hey, new idea, new concept, something like more in that direction. There are new projects, of course.

I'm not even 100% sure this was a marketing thing, but outside of the venue on the morning of the first, maybe the first day, it was the morning of day zero. I saw flyers posted and it wasn't on venue property. It was like on the streets that kind of outside the venue. And they were white, like letter size sheets with black inkjet looking prints, very plain, like looks like handmade prints.

and it had a big QR code on it. And it said, "Jason, I know you are cheating on me with Mandy. I have proof. Pictures don't lie." And then the huge QR code. And I would bet money that somebody's marketing.

I still don't understand how you can't scan the code. Because I like the mystery. I didn't scan the code. Because either way, if it's marketing, I don't want to give them the counts. And I don't know if someone's public airing their own dirty laundry. I don't want to encourage that either. I just want to talk about it because I think it's hilarious.

If it is marketing, that was probably one of the strangest because I don't scan QR codes. So there would be a 0% chance I would even, whether it was something I wanted to see or not.

What if you go to a restaurant and you have to scan a QR code to see the menu? Do you not eat? I don't go to that restaurant. Okay, so let me tell you this. If you ever go to China, you will not return. You will die after hunger. Starvation. I would have a burner phone then, so it would be fine. I would probably scan then because it wouldn't be tied to anything real. Did I say that out loud? I think I did. Okay.

So we can keep this short because it sounds like it was just... It sounds like we have no idea what just happened. I can tell you all sorts of little things that charmed me or whatever. Here's just something interesting. Heroku open-sourced 12-factor app 15 years later. Wait, open-sourced the text? I guess so, yeah. So the web page is open-sourced now? Mm-hmm.

Why? Why? That's a great question. That's one of the things that I pulled out for next week's or this Friday's live stream is Heroku open source 12-factor text.

Okay. So, yeah, okay. 15 years later. What else happened 15 years ago? You know what would be cool to check that page, assuming that they didn't come and change it? Maybe it's one of those HTML pages that is sable-scoping the text. And now we can. I don't know. What were the other things that you saw that you thought, oh, that was interesting?

There's a new project called Flat Car. Do you know about that one? It's an open source container Linux. Oh, well, they talked about it on the keynote stage like it was new. Oh, yeah. It's been around for at least a year. Oh. Well, I feel the second time they got me, I'm gullible. Yeah.

Okay, but if it's been around for a year, that counts as new. It's still new, yeah. I saw a talk at Rejects that charmed me a lot, and that was Duffy Cooley. Oh, I wrote down the other guy's name. Duffy Cooley and Kyle Quest, and they did a talk about malicious compliance as far as scanning for vulnerabilities.

And so the first part of it was like using Docker Slam or Slam AI or one of those things. Like you scan your image, you have 4,000 vulnerabilities, you need to release in 24 hours, what do you do? So the first one was like slimming down the container image to the bare minimum and we cut those vulnerabilities in half. But then the second one was like a tool that just obfuscated the metadata. And then it was like zero vulnerabilities. So that's it.

If you cannot get rid of vulnerabilities, you can trick vulnerability scanners to... Yeah. So they don't discover it. Yeah. It sounds brilliant to me. It sounds like something... Actually, we have a tool for what we were doing in the past already.

The whole world is just built on bubblegum and duct tape. You just need to hide the metadata and you can trick all of the vulnerabilities in it. I don't know. It should be horrifying instead of funny, maybe. I was once involved in a project where we were supposed to remove all high-level vulnerabilities.

And then comes, I don't know, a few days before go live and they're still there. And that's one of the requirements. But we had the flexibility to actually upgrade or downgrade vulnerability. So we just downgraded all the vulnerabilities to medium or something like that and we were done. Yeah.

Now, I know, Whitney, you went to Rejects. Victor, did you go to Rejects? And did you all do something at Rejects? Did you do anything together? I was emceeing. Oh, you were emceeing. I was excellent at it. Everybody congratulated me afterwards. I mean, my emceeing was mostly world start. And you're done, go out. I was congratulating you on time. Oh, I was keeping you on time. Your time is out, man. You're out. I don't care.

I took it from Whitney's playbook when we do your trues. You actually do get to physically push someone off the stage this time. It was Reject's

They do such a good job. It's free now, which is cool. And they pick really nice spaces that are just wonderful to be at. And the community is so good, and you get a chance. It's just such a good, good, good thing. Yeah. I think that anybody coming to KubeCon and skipping Rejects is making a terrible mistake. Yeah. And the hallway track at Rejects is wildly valuable for anybody, not just for those in our industry, too.

And we're also happy to see each other and we're not tired yet and we want to talk about everything and take advantage. So what you're saying is you use up all your energy at Rejects and don't save any for the actual conference itself. It's a slope. You know how, for example, if you're about to do a marathon, you don't just start running.

First you need to stretch and maybe jog for a while, warm up and things like that, right? To me that's rejects. It's kind of, it's a preparation for the marathon. And there's lots of hope in the air. Yeah, lots of excitement. It's electric. I'll tell you something else, the platform building space, like there's a lot more about platform engineering this time.

I think that Platform Engineering Day or whatever it was called was the biggest one. It was the biggest one and they're going to have two tracks next KubeCon. Exactly. So that's going to be, as far as I know, the first Day Zero event that is in two tracks. Yeah.

And they have platform coffee every morning. And I went to platform, so it's like it's 7 a.m. every morning. This platform, people want to talk about platform engineering, I'll go to the same coffee shop. And on the first day, the poor people worked at a coffee shop. We tried to call ahead and say we were coming, but like 50 people came in at once and absolutely overwhelmed the staff.

But I was, I went to platform coffees last in Europe in Paris and there are maybe 20 people. And so it's more than doubled in size. And it was really cool too. And all four days people showed up. The last day had more like 25 rather than 50, but still impressive after four days of madness.

So, Victor, I remember last time you were signing books and the hall was yelling at you, like, you got to get the people out of here. I assumed you signed books again here. Yeah, yeah. I think they brought 100 copies and that was gone very, very fast. Oh, so you only had 100 copies. So that was easy. Yeah. You didn't have to wait too long. Okay. Yeah. And Victor and I gave a choose your own adventure talk. And it was...

It was really, really funny. We had a really good time. The video's already live, which is a thing that is really impressive. It was like live maybe the same day. Okay. And...

It's the usual thing where I talk about it. It's observability odyssey. So I talk about what is the system design choice that we're making a decision about? What are all the tools that can do that thing? And then we give a live vote to the audience and they choose which tool they want to see Victor implement in the ongoing demo. And there's a...

We're on a race against time to get it all done before our session time elapses. And we were down to the wire. We were counting 10, 9, 8, 7, Victor Lettick scrambling to do the last bit of the demo. And I won't tell you whether he made it. You have to watch the video. But there was lots of cheers and laughing. And overall, it was just one of the funnest moments of coupon for me.

And I think for everyone in that room. So let's go ahead and look ahead to next year. 2025 was posted. London is April 1st through 4th. China is June 10th and 11th. They're adding Japan officially. That's June 16th and 17th. So you can go to Hong Kong and then you can go up to Tokyo. Just back to back. Okay. I didn't know. They're back to back. Okay. That's cool. Yeah. They are...

whatever the 10th and 11th so it's like within seven days you're in hong kong and tokyo both so hyderabad is august 6th and 7th in india and then next year in north america it's atlanta from november 10 through 13. number one it's a shorter flight for victor it's probably direct so no no connections oh yeah that would be nice that will be amazing

Whitney, for you, it'll be easy. London will be easy for both everybody. It's just going to be Hong Kong and Tokyo. What are you going to do about that? I mean, I'm not yet sure whether I'll go to Hong Kong and Tokyo, to be honest. So there is a fundamental difference in how CNCF approaches KubeCon Europe, US and Asia on the other hand, right?

US and Europe are supposed to be one big event for the whole continent, right? So people from all countries in Europe are going to be in London, people from all parts of US plus Canada and what's about in Atlanta. But for one reason or another that does not work in Asia, right? The previous KubeCon failed to attract people from outside the country where the KubeCon is.

So that's why they're organizing China, Japan, India, right? And which is awesome, right? It's probably great that, hey, you can have KubeCon in your country instead of dealing with visas and money and what's not to go to another one.

But now from the speaker perspective, it's a really, really long trip. And those are relatively small events. I would say that they're actually on a level of a KCD in a way, right? Maybe a bit bigger, but not significantly. Like 1,500. Yeah. I mean, 1,500 is nothing definitely, right? But not 9,500 that we just experienced. Yeah, exactly. Exactly. So I'm not saying I'm not going, but

It's nowhere near the priority list as KubeCon in the U.S. and Europe. Now, they've also gone ahead and announced for 2026 as well, at least Europe and North America. March 23rd through 26th in 2026 is Amsterdam. Yeah. North America is going to be October. So moving up from November to October in Los Angeles, California. Yeah.

That just seems miserable to me. Why? LA? LA is not one of my favorite cities. I like LA, but it's not one of my favorite cities. True, but from the logistical perspective, LA probably makes more sense than Salt Lake City. I'm not now talking from the how much you like or dislike the city, but logistically it's probably... I feel that part of the reason, probably not the big one, that there is a lower attendance in the US than Europe could be location, right?

Because more people might be inclined going to LA than Salt Lake City or what was the previous one? Detroit, Chicago. Detroit, Chicago, right? Yeah. One thing, I thought I hadn't been doing this too long, but someone asked me how many KUKONs I've gone to, and this is my ninth, so I guess I have been around longer than I realized. But Amsterdam and LA will be the first time I'm doing repeat KUKON cities. Wow.

Right, because how long has it been in LA? It's been... That's like a long time. Or am I confusing that with Doc O'Connor? 2021? Yeah, COVID. Yeah, 2021, I think. Yeah, it was 2021, definitely. I mean, I don't even count that as Cape Cod, to be honest. Yeah, well, I didn't know better. I didn't know anything. It was my first little deer in the headlights. So during the keynotes on day three, which was the best day to go to keynotes because it's all about the community.

At the end, what they did, they had Tim Harkin host a Family Feud-style show where people from the community, like it really had a Family Feud music. They had the X's and the ding when they got it wrong. So are you both familiar with Family Feud? Not great. Okay.

I am. Okay. So with Family Feud, they do a survey and then you need to guess who answers. So for example, under Kubernetes context, the survey question is, what is the biggest risk to Kubernetes and cloud native?

And so Tim Hawkins somehow conducted a survey of the CNCF community and asked them this question. And then the contestants have to guess what are the top five answers to this. And whoever gets it right in the right way gets points. It was very fun and funny. It was partly fun and funny in the way Victor is fun and funny because the person who was like,

controlling it was really messing up the controls and just so it's partly funny because it wasn't super polished but it was adorable but I thought it would be fun if I asked you to okay to come up with their five top their top five things maybe come up with two of them okay what are the biggest risks to Kubernetes and cloud data what is the Kubernetes community think is the biggest risk to Kubernetes and cloud native

Five things. I thought you were going to give us five things and we choose two. No, you need to think of something. Okay. Complexity? Yes, that's the second most popular one. Good job. Do I get a prize? Think of two or three of them. One more. Okay. I let Darin answer while I think of the second one.

the biggest risk to Kubernetes and cloud native. Not using it. The big companies, the

-Cypress Killers. -Oh, Cloud Provider Lock-in is on the list. Yes, good job. -Okay, there we go. -Yeah, yeah. -Oh, I'm killing this game. -Do you want me to tell you the rest of them? -Yeah, go for it. -This is the community said, "Lack of innovation," which I don't really understand, honestly. AI. -Okay. -And lack of maintainers. -What was number one? -The most popular one is lack of innovation.

The second most is complexity. Third most is cloud provider lock-in. Fourth most is AI. And fifth most is lack of maintainers. Boy, I would have to flip that around to lack of maintainers up to number one. Yeah. People are getting tired. It's true. And those patent trolls, those evil patent trolls, what is the best feature in Kubernetes? There are six top answers.

API? I mean CRDs, API, accessibility. Okay, custom resources, yes. Number one, yes. Good job. See, killing it. Killing it. The best feature in Kubernetes. Standardization. Yeah, controllers and declarative. I'll give you standardization for that. Yeah, that's number three. Okay.

You said two. I'm done. Okay. I can be done. Do you want to guess more, Darren? No, I don't want to guess any. Okay. The most popular is custom resources. Because he won number one. And the second most popular is the control plate. The third most popular is controllers and declarative. Fourth most popular is pods. Fifth most popular is the community. And sixth most popular is auto scaling. Okay. And then we have one more. Do you want to do one more or are you tired of this game?

Do one more. If there's only one, then we can do one more. Don't hide your enthusiasm. That's my polite way of saying, yes, I'm tired, but let's go ahead and do one more. Let's indulge you anyways. Name a barrier of adoption to Kubernetes that has largely been overcome. That has been overcome? Yes. Setup.

Or installation or... Cloud provider support, maybe. We'll call that. Yeah, that's managed Kubernetes cluster. Or actually, it's also related to number one, which is cluster lifecycle. Yeah. Yeah, cluster lifecycle. Cloud provider support is number two. Number three is hiring. Number four is reliability. And number five is documentation.

Good job, everyone. Thank you for indulging me. No, that's fine. And I forgot that you normally put all those kinds of things together. Yeah, and wait until Darren starts with his chopping board. All that is gone from the final one. That's fine because I'll never listen to it. I would have never known. In my mind, it lives forever. There you go.

Well, Victor Whitney, have a safe trip home. Enjoy your last few minutes, I guess, in Salt Lake City. What's the big plans for dinner tonight? I'm going to some Italian and she claims she's not eating ever again. I'm flying my best friend out from Texas to come hang out with me in Salt Lake City. We're going to do some of the nature things, drive out to the salt flats. So I think I'm going to sleep tonight so that tomorrow when my friend arrives, I am some sort of human. I can...

perform some friendship duties as opposed to like flat on the ground face down all day. Maybe that's the title. This is some sort of human. That's, that's how that should be the tagline now for all cube cons, some sort of human, or maybe that becomes AI again. I don't know. Well, y'all have a good trip home and we'll talk to you again soon. Thank you. We hope this episode was helpful to you.

If you want to discuss it or ask a question, please reach out to us. Our contact information and a link to the Slack workspace are at devopsparadox.com slash contact. If you subscribe through Apple Podcasts, be sure to leave us a review there. That helps other people discover this podcast. Go sign up right now at devopsparadox.com to receive an email whenever we drop the latest episode. Thank you for listening to DevOps Paradox.

We'll be right back.