Mike Wagner is the Co-Founder & CEO of Metify.io. He joins the show to talk about his company, career experience working with data centers, and infrastructure. He also tackles the bare metal servers and how this is being utilized in his company. Moreover, they dive into Metify's infrastructure and operations tool for managing and provisioning servers, and many more! 

Will:

Welcome everyone to another episode of Adventures in DevOps. And it feels like I haven't been here in forever. If you don't know who I am, I'm Will Button. I am the co-host here. And joining me in the studio today is Jonathan Hall.

 

Jonathan:

Hey everybody, how's it going?

 

Will:

Good, good. And then we have our guest today, Mike Wagner. How are you, Mike?

 

Mike Wagner:

I'm doing awesome. Thanks Will.

 

Will:

Right on.

 

Mike Wagner:

Hello Jonathan.

 

Will:

It's great to have you here, because today we're going to be talking about some bare metal stuff, right?

 

Jonathan:

Oh, I thought it was heavy metal. I thought we were doing all rock and roll today.

 

Mike Wagner:

We can do a little of both.

 

Will:

Oh,

 

Mike Wagner:

I'm

 

Will:

I-

 

Mike Wagner:

always like, yeah.

 

Will:

my bad, I thought bare metal was like a subgenre of heavy metal. You know, cause

 

Jonathan:

Oh.

 

Will:

you have like death

 

Mike Wagner:

I'm sorry.

 

Will:

metal and emo metal and goth metal. I thought bare metal was another... Okay.

 

Jonathan:

Is that where you play guitar naked? No.

 

Mike Wagner:

I think it is. I think it is. I'm sure that there's a substack dedicated to it.

 

Will:

And for everyone who's

 

Jonathan:

There was

 

Will:

watching

 

Jonathan:

a Reddit

 

Will:

on

 

Jonathan:

group,

 

Will:

YouTube,

 

Jonathan:

but it's on strike

 

Will:

right?

 

Jonathan:

right now.

 

Mike Wagner:

A community, a subsack, a YouTube channel. I'm sure we've just scratched the surface of something amazing.

 

Will:

Right on.

 

Jonathan:

So what is bare metal? I mean, we've all at least been involved in some way, but something is not forefront of our minds these days. What is it?

 

Mike Wagner:

Yeah, so a bare metal server is just a server that you generally run a single instance on. So it's a dedicated server. Okay, so you're, whoever is, and it's different things depending on what level you're approaching it, but from an infrastructure and operations perspective, we like working with enterprise architects, VPs of IT, the guys that are responsible for. creating the clouds, the private clouds, the hybrid clouds, whatever it may be. So across the board, we work with all of the big OEM players. So it's Dell, HP, Supermicro, Lenovo, Cisco. And, you know, whenever you talk about, you know, what's the cloud or what is the edge, you know, the answer is truly it's just someone else's computer, right? That fabulous meme. that's been around for a while now. So somebody has to work with those servers. Somebody has to keep the BIOS up to date. Somebody has to keep the firmware of all the peripherals where they need them. And it was kind of a neglected segment, this whole idea of low level operations and working with bare metal, working with servers. So yeah, that's, I guess, in a long winded way. Bare metal is just someone else's server.

 

Jonathan:

That's what I thought the cloud was.

 

Mike Wagner:

Yeah. Well, the cloud's based on bare metal.

 

Jonathan:

The

 

Mike Wagner:

So

 

Jonathan:

cloud is made of metal, wow. We're in

 

Mike Wagner:

yeah,

 

Jonathan:

some dense atmosphere here.

 

Mike Wagner:

isn't that amazing? And now of course they're reselling that as, you know, bare metal as a service, right? So you can, uh, you can actually get your servers, uh, just the way you want them sort of, um, a custom metal, if you will. Um, and wow, the premium on that is really amazing to witness. So what we've, what we recognize my CTO and I and co-founder Ian Evans, Years ago, we recognized that there was this almost everyone sort of getting drunk on public cloud and virtualization. And there's an efficiency gap. There's an architectural problem with it. And then there's also security and all sorts of other concerns kind of that we recognize that if we could find a way to make it easy to do DIY bare metal, so you could build your own clouds easily, that almost public cloud-like ease. And

 

Jonathan:

Mm-hmm.

 

Mike Wagner:

on your last episode, I enjoyed the GCP versus AWS discussion.

 

Jonathan:

Mm-hmm.

 

Mike Wagner:

And I must say I fall in the GCP camp as well. I seem to have kicked tires on AWS multiple times and just paid for something that didn't really lead to a lot, but, um, GCP has always worked fairly efficiently. Um, so yeah, so it's, it's the idea. The idea was to make something, um, so simple, you know, just to create a UI that made it incredibly easy to, uh, discover provision. and maintain your servers across multiple platforms. So any hardware provider, any hardware manufacturer through a single pane of glass.

 

Jonathan:

Nice.

 

Will:

Right on. So there was a thing came out by the guy who did Ruby on Rails, DHH. I can't remember what the DHH stands for, but anyway, he was talking about moving his company from the cloud to bare metal and the savings that they were forecasting to save that. And I think in his particular case, we talked about that on one of our previous episodes, it makes sense. Is that the... people that you're trying to target there as individuals who are trying to reduce their spend and gain more control over their environment.

 

Mike Wagner:

Yeah, so that's certainly, so cloud repatriation is the buzzword

 

Will:

Oh, that's

 

Mike Wagner:

of the

 

Will:

a

 

Mike Wagner:

day

 

Will:

nice

 

Mike Wagner:

for that.

 

Will:

term. I like

 

Mike Wagner:

Isn't

 

Will:

that.

 

Mike Wagner:

that nice? There's a lot of repatriating going on lately and we're working on most parts of it

 

Will:

Yeah.

 

Mike Wagner:

being the patriots that we are because we actually are, love America, have traveled the fair amount and man, the United States is awesome, but it's another story. So cloud repatriation in Yes, there's organizations that you reach a certain point in your cloud spend where you can save a lot of money by bringing it in-house. There's a lot of factors related to that. So having the right skills, having the right tools in place, those are all really critical to make it happen. And we make that ridiculously simple. So yes, we've done repatriation projects and... you know, the results are always awesome. I mean, so far, everything has worked out very well, I should say, because, you know, I'm sure there's large projects out there and enterprises that could tell horror stories. But as long as you have the right tools in place and you don't try and bite off too much at once, it's similar to cloud migrations, really, right? And then, you know, everybody shoved stuff up into the cloud and then they realized when they got the bill that maybe this was dangerous. And then

 

Will:

Great.

 

Mike Wagner:

of course, there's always the problem of shadow IT, right? You know, everyone, you know, popping up, uh, AWS or GCP instances and, you know, nobody being able to really track what's out there. So, um, yeah, no, it's, uh, we, we definitely do some cloud repatriation. We do a fair amount of white space, uh, where it's just, okay, we have a new project, we know we want to do it, um, in house and a lot of hybrid as well. And then the edge stuff, that's the thing that, um, you know, really played in our favor was just the, uh, expansion of the architecture. and of where compute is being placed now. Because of the ever shrinking platforms you can put tremendous compute on. So Moore's Law and the really fun march of technology getting smaller and faster and just more amazing and being a tech geek, it's just a blast to get to work with these small form factors. And so we've done some really cool stuff with the DoD related to that and some ISPs as well. some edge, I guess I shouldn't call them, some edge compute players overall. So yeah, the space is exciting. There's a lot happening in it. And being able to provide a platform that can make that easy has always been our overall vision.

 

Will:

Right on. Whenever it comes to actual physical location of these servers, do you have the space that you provide or do you work with people with their existing spaces or what's that part look like?

 

Mike Wagner:

Yeah, so we have both. So we work with a number of, excuse me, hosters, colos, data centers that we can partner with. And that's the other thing, right? We're almost exclusively a channel-led company. So we work with partners across the board. So SIs that do OpenShift implementations, a managed service practice or have a colo facility or colo partner. We work with colos directly. So yeah, a little bit of all of it and

 

Will:

Okay.

 

Mike Wagner:

depending on what the customer wants. The big thing, one of the big things that we recognize also back when we were sort of alpha testing the product in our prior lives before we launched as a startup was that the Fortune 1000 set. They've got, you know, they've been in place for a long time, usually, and they have very particular tools that they like using. They have particular architectures that they like. We wanted to give them the freedom to choose and stay on whatever they were working with. We would provide guidance and really are largely preference-free because we've, depending on the use case, it's definitely each implementation is, its own sort of beautiful snowflake. You know, it's it is what it is. And a lot of the enterprises have done excellent work establishing the practices that they have in place. And this just, you know, we let them keep the choices that they want. And then we come in with this low level platform that makes, you know, handling the metal itself incredibly simple. So. So

 

Will:

Right

 

Mike Wagner:

yeah,

 

Will:

on, yeah. I think

 

Mike Wagner:

little

 

Will:

that

 

Mike Wagner:

bit

 

Will:

makes

 

Mike Wagner:

of

 

Will:

a lot of

 

Mike Wagner:

little

 

Will:

sense.

 

Mike Wagner:

bit of all of the above.

 

Will:

Yeah, because whenever you come to, whenever you think about bare metal versus cloud, in the cloud you choose from whatever product that cloud provider is offering you, but in bare metal, it's like, hey, here's a CPU, some compute, some memory, go for it.

 

Mike Wagner:

Yeah, yeah, absolutely. And now custom chips and GPUs and this, your chat GPT episode was really cool. I mean, just all the things that are happening on the edge or wherever with chat GPT-4 and the other AI players, the demands around what we're trying to do from a compute perspective have just gone through the moon. you know, now it's a matter of, okay, where can you get access to these advanced chip sets? And where can you get access to the bare metal to do the crunching that you want it to do? So it's a really fun space to be in right now.

 

Will:

Oh for sure.

 

Jonathan:

What are some of the biggest differences when you're dealing with bare metal, from your standpoint, not from the customer standpoint? Because I imagine you try to make as much of that transparent to the customer as possible. You get advantages without the extra toil of dealing. They're not physically installing hard drives and running cables. From your perspective though, managing this stuff, what are the differences between selling bare metal services and if you were just selling traditional cloud services?

 

Mike Wagner:

Okay, yeah, so we don't personally sell it as a service, if you will,

 

Jonathan:

Mm-hmm.

 

Mike Wagner:

but what we've seen from an implementation perspective with regards to how we handle different hardware footprints, that's sort of its own beast for us, right? What we do wasn't possible five years ago because the standards weren't there. So these low-level hardware standards just... Have really become ubiquitous in the last two to three years and in particular the folks at the DMTF have done a great job Distributed management task force that came out with the Redfish Specification so we'll get geeky for a second here. And you know, these are the low-level guys that Have put the open project in place that Essentially everyone has gotten behind that or open VMC, which is part of the open compute project And between those two, as well as a few other open standards that are out there and driving change, like SNIA has Swordfish and there's a Yang to Redfish model as well. There's a number of open standard projects that are going on that are all sort of working as one now, if you will, with sort of Redfish as the almost underlying base foundation of an open API to write towards. things have really progressed for us. So how we work with other, with each OEM, that's kind of part of our secret sauce for sure. And not everyone adheres to the Redfish specification perfectly, which is, which has always kind of been one of the problems with open standards. You

 

Will:

What?

 

Mike Wagner:

know,

 

Will:

Oh my god.

 

Mike Wagner:

yeah, imagine that.

 

Jonathan:

I thought only

 

Mike Wagner:

So,

 

Jonathan:

Microsoft had that problem.

 

Will:

Hahaha

 

Mike Wagner:

yeah, no. Especially the guys in the cellular space. Wow, are there war stories about that from an open standards perspective? We've done a fair amount in the radio access network space, in the RAN space. They love their acronyms more than any other, the Teleco

 

Jonathan:

Hehehe.

 

Mike Wagner:

guys. The open standards debacles that went on there, and still are, just kind of wild and crazy to witness. Finally, on the data center side, things have really, really come along in the last couple of years. So all the manufacturers now of the motherboards are on board. So all of the servers that go out, essentially, will have a BMC that adheres to one of the other specifications or both. And yeah, you can have at it. And with some tweaking, with some customizations, you can create that single environment to work on any server. from anywhere, you know, so that's really the cool thing. As long as there's a plug going into that server and a network connection of some type, we can do, it's true out of band lights out management, right? So it's that uptick in service that you normally had to pay a lot for if you wanted to do it in a proprietary way. And then you'd have to learn each proprietary tool and we got rid of all that. Just one tool to rule them all. if you will.

 

Jonathan:

the

 

Will:

Wow, that's crazy. It's been a long time since I've dealt with bare metal. And to be honest, I haven't missed it. But it sounds like it's come really a long way in terms of being able to do that out of band management and get the visibility of what's actually going on at the hardware level. And so this is a service that you provide as you take the hardware and you integrate all of that so that the customer has that visibility remotely to the physical hardware.

 

Mike Wagner:

Yeah, yeah, it's a down. So our software tool, right? It's a kind of viewed as, you know, an out of band utility platform. Okay, so and you can you can download it and check it out for 30 days. And, and then yeah, immediately start discovering your assets. So that's the thing, you know, the other big piece, it's it does a discovery. And then we've got a service catalog that's built in. And you can, you know, provision whatever OS you want. You can go up stack with that and provision, you know, an OpenShift cluster or an Anthos cluster, a Rancher cluster, we work with all of them. So it's a powerful tool. And with great power comes great responsibility. So we take security very seriously and we've air gapped multiple networks internally within the product to make sure it's safe. And... adheres to some NIST and DOD certifications. So there's, you know, across the board, we kind of wanted to make this enterprise ready and address the big concerns that we knew our enterprise customers had going into this project, going into if whenever they want to take on a bare metal project, there's the number of things that, you know, Ian and I both had. a lot of years working in data center environments. And he especially had to fix a lot of issues and architect and design things from the ground up. So he took all that experience and kind of, you know, put it all into sort of the dream I&O product, the dream low level utility to make life easier for data center operators.

 

Will:

Right on.

 

Mike Wagner:

Yeah.

 

Will:

So I'm gonna show, possibly show my ignorance here. If not, it'll become evident by the end of the episode.

 

Mike Wagner:

I

 

Will:

Is SNMP still like a thing you mentioned, the auto discovery and the service discovery of the hardware? And I remember

 

Mike Wagner:

Mm-hmm.

 

Will:

long ago using SNMP to try and accomplish that. Is that still the pattern?

 

Mike Wagner:

Yeah, well, you know, SNMP is still out there. I think, you know, and this is something that Ian could answer much better than me, but we've, so the key now is we're all restful. So we switched to rest. And that's where obviously the industry, right, API and API for everything sort of idea.

 

Will:

Mm-hmm.

 

Mike Wagner:

And you know, the SNMP is a protocol, certainly. Is still out there and I'm not sure how often we interact with it because the availability of all of the Pieces that we need are essentially there via rest and via you know A standardized API the Redfish specification

 

Will:

Oh wow.

 

Mike Wagner:

the Redfish API itself. Yeah So it's a big shift in how the underlying infrastructure can be discovered and worked with.

 

Will:

Right

 

Mike Wagner:

Modern,

 

Will:

on.

 

Mike Wagner:

the big modern shift. Yeah.

 

Will:

Yeah, for sure. And I hadn't really thought about this problem in the past, but there was a ton of opportunity there waiting just in that level of abstraction. So yeah, I can see the thought process and why you would tackle this problem and the potential improvements you can make on it.

 

Mike Wagner:

Yeah, it was really cool. It's one of those when we showed it to customers initially, when we showed to them now, but even back when we were outplaying it, the response was consistent. It's like, okay, cool, let's get it into our lab. We like it. We'll buy it. There was no product. And the one, the bit of guidance that we got was we wanted to be, we want no sort of proprietary pieces in it at all. We just want it to be not beholden to any particular software stack. or hardware provider, we just want it based on open. So that really required us to launch it as a starter. But as you're saying, you can picture it, how it's helpful. As soon as we showed it to these VPs of IT, these enterprise architects, they're like, oh my God, okay, you did it. I had this idea. So we got to hear a lot of sort of anecdotal stuff about how they wish this was in place. If this was only there when I was at eBay, Yeah, some tons of stories, right?

 

Will:

Oh yeah.

 

Mike Wagner:

And, and yeah, you know, the crazy part about, it's just somebody else's computer is, it has become so simple to just fire up an instance, but at scale, you know, the bills are insane. And then, you know, with the, with the power of compute today, the cost savings as well as the security considerations, you know, AWS has been, handing over a lot of information that companies didn't want handed over. So there's security, specific security profiles that we work with that, as well as sort of, what's the best way to put it? So high value targets, if you will. So we have some

 

Will:

Mm-hmm.

 

Mike Wagner:

folks that are well known and they wanna make sure their stuff is secure. So another use case for having private cloud is just knowing that it can't be turned over to, you know via subpoena without your approval right so um yeah there's really interesting just tons of use cases out there um that uh that bring a need for this sort of immediately recognizable and uh yeah the guys you know when they see it they're just like oh god you know there it is finally so

 

Will:

Yeah, for sure. I think that's a lesson a lot of companies in the last couple of years are learning the hard way that the cloud providers and in addition to that, SaaS providers were very convenient to onboard, but they're also very willing and ready to turn over your data to anyone who asks for it.

 

Mike Wagner:

Yeah, yeah. Well, it's another revenue stream. I mean, you know, it's

 

Will:

Oh

 

Mike Wagner:

stunning,

 

Will:

for sure.

 

Mike Wagner:

right? It's yeah, if you don't check the paperwork carefully, watch what you sign up for. Click through contracts are who you just never know. So yeah, no, that's a that's a great point. It's it we're in a wild and data we're in the you know the data wild west, if you will, anything they can grab they will

 

Will:

So what's a typical

 

Jonathan:

What kinds

 

Will:

onboarding?

 

Jonathan:

of...

 

Will:

Go ahead

 

Jonathan:

No,

 

Will:

Jonathan.

 

Jonathan:

you go first, Will.

 

Will:

Okay, I was going to ask what's a typical onboarding or implementation for your software look like? What's the barrier to entry?

 

Mike Wagner:

Yeah, it's a very low barrier to entry. We're about to send out our latest release. We'll allow for single click download and installation. So we're really excited about that. We're probably three weeks away from getting that out. So it's really simple. You can download it and then we can walk you through sort of a white glove installation process to make sure everything's working right. And that's done in a couple hours. Or we have an appliance as well. And we have a few customers that just said, cool, build us an appliance and ship it out. So we've got, and you can pick the form factor on that as well. So we've got palm tops, we've got two use. single use, whatever the use case calls for, where as flexible as our customers need us to be. Because given the locations, the points of presence of some of these compute environments, it can be something that we need to be way about a pound and fit in the palm of your hand versus a two-year, pretty heavy horsepower. couple of servers. So, yeah, but overall the software itself is very easy to download and to load and work. And like I said, I'm just a couple of weeks away from making it a single button go.

 

Will:

Right on.

 

Mike Wagner:

Yeah.

 

Will:

So that should be about the time this episode releases. What's your website?

 

Mike Wagner:

Nice. Yeah, it's a www.metify.io.

 

Jonathan:

So I'm curious, what types of clients, I don't want you to tell us names unless

 

Mike Wagner:

Mm-hmm.

 

Jonathan:

you have agreements that allow you to do that, but like what kinds of clients and what kinds of workloads is this ideal for? And the second part of the question is, who should not do this?

 

Mike Wagner:

Hmm. So yeah, it's a fun one for us because, well, OK, so a number of our customers we have, obviously, NDAs in place with and they don't like us. They feel there's a competitive advantage related to how and who and where and all that good stuff. They are implementing Mojo as part of their architecture. Having said that, Major League Baseball published a really cool article. as well as in their technology blog about how they use Mojo Platform to power all of the baseball stadiums in North America. So as far as use cases go, this is kind of the perfect hybrid edge use case that has GPU dependencies and hybrid in that 7.2 terabytes of data per game are then uploaded into GCP. and it's a Anthos push button Anthos implementation. So yeah, this is, you know, from a use case perspective, we did not think Major League Baseball was going to be, we didn't envision them as a customer, but it makes sense, right? And when you're as low level as we are, it's almost like asking Intel, you know, what's your perfect customer, you know? It's pretty much anyone that does is compute at scale. And has the need to you know access these servers and make sure that the bios is up-to-date Make sure that you know any zero-day vulnerabilities can be taken care of without having to fly in and do a thumb drive shuffle or as Kevin Backman from Major League Baseball calls it the handjam You know and that's you know, just travel and maintenance expenses on servers because it is What underlies private clouds public clouds, you know, it's a significant budget hit.

 

Jonathan:

Mm-hmm.

 

Mike Wagner:

So yeah, which is another reason why public cloud was as popular is as popular as it as it is because, you know, okay, well, you don't have to worry about that. Right. Well, you know, similarly with this, it takes away the concern of having to fly somebody in and update the bios or update a firmware. And, you know, remotely monitor exactly the health of that system or even create a snapshot. and then make sure from a governance perspective that there's no configuration drift on that server versus the rest of the servers in the field. So we've got customers with thousands and thousands of servers, and with configuration drift, something can happen, right? The guys in the labs push out an update, and it's possible that if they're not exactly aware of the BIOS levels and what is in those what chipsets are in those servers that it could cause something to go down. And you know, bricking a subnet, bricking a major portion of a company's intranet is a bad thing. So yeah, we make that a lot more governable and transparent to the guys that need to know that information.

 

Will:

That's crazy. I hadn't even thought about that use case, but like baseball stadiums, you know, where they have a huge compute infrastructure just in the building there and financial as well like bank networks that have networks of ATMs. I mean, we've

 

Mike Wagner:

Yeah.

 

Will:

all seen the blue screens of death from an ATM that's still running Windows XP and stuff like that. And it seems like this would be a would fall into your use case of things to manage.

 

Mike Wagner:

Yeah, for sure. No, and we're working with a few banks. We're just, you know, scratching the surface on retail. We think that's actually going to be one of our largest verticals. And, you know, surprisingly we're actually not in production yet with any major retail player. Well, it depends on how you consider MLB, but in terms of some of the use cases that we are. in place for them. But yeah, you're absolutely right. It's one of those things where in fact, early on in my networking career at IBM, I had to go around and do a number of updates for a bank and flew around the country with a set of disks and a hard drive. It was unbelievable. process, where I was flying around from city to city, could easily be done. And we have done it from a single location and doing all of the same functions I had to do, which was cold booting and formatting and all the things, the new OS loads and the BIOS updates and all of those things. So it is really cool to see the advancement of where we are now. and all of the cost optimization, as well as the architecture optimizations that can occur because of it. Because the other big piece of this, and this is, you know, adventures in DevOps-y, Kubernetes and containerization in general, right, dramatically changed how and what the optimal state of applications is to operate in. And... So where VMs are heavy and actually performance reductive, you can get containers on bare metal and really have a screaming application with outstanding security. And so yeah, the use case piece as it pertains to containers and applications and how they're presented architecturally to developers. That has really driven a lot of this as well.

 

Will:

I remember something similar back pre Y2K. I was working in the telco space and same thing flew all around the country, updating a bunch of telephone systems for Y2K, carrying the floppy disk with me, upgrading them.

 

Mike Wagner:

You said Y2K.

 

Will:

And

 

Mike Wagner:

That's awesome.

 

Will:

I did, I did.

 

Jonathan:

Ha ha

 

Will:

And

 

Jonathan:

ha.

 

Will:

I'm willing

 

Mike Wagner:

Yeah.

 

Will:

to bet that a significant percentage of those systems I updated for Y2K, that's probably the last time that they were updated because of how big of

 

Jonathan:

Mm-hmm.

 

Will:

a hassle that was.

 

Mike Wagner:

Yeah, well, I mean, yeah, telco is we love the space and it is changing quickly, which is a, you know, kind of a oxymoron to even say because nothing in telco changes quickly.

 

Will:

Hahaha!

 

Mike Wagner:

Their approval process can take years. But, but yeah, those van rolls. And, you know, it's funny is I'd be willing to guess that a number of folks Gen Zs, they have no idea what Y2K is. You might have to do

 

Jonathan:

Hehehe

 

Mike Wagner:

a postscript on this podcast

 

Will:

Fair

 

Mike Wagner:

just

 

Will:

point.

 

Mike Wagner:

to explain what Y2K was.

 

Jonathan:

We're a bunch of old guys.

 

Mike Wagner:

Yeah, man. So, yeah, there is the need, which is just amazing from the amount of van rolls that still have to occur. the amount of flights that still have to occur. So there's a green element to this, you know, as well. It's like, and then we do have a green data solution. We're working like with lead compliant certifications and, you know, lowering carbon footprint as much as possible through the new schemas that are being put in place through the Redfish. So you can monitor thermals, monitor temperature in rows and in racks and in chassis. So all the cool new sensor stuff that's coming as well. We've got a really cool green data center use case that we've built out as well. So yeah, there's a number of very pertinent problems that never got solved, but the tools are there. It's kind of like the ring doorbell guy. It's like, hey, if you put a few of these things together, you got a pretty cool solution. And it's kind of similar to that except. It also took open standards and everyone to play nice for it to really happen.

 

Will:

For sure, you mentioned that the standards have come a long way. Are you seeing a lot of enthusiasm? Because it's one thing everyone's like, oh, OK, we'll support this. But then to be enthusiastic where they're like, no, we're really going to prioritize this. Are the hardware manufacturers and OEMs really seeing the value of what you're doing here?

 

Mike Wagner:

Absolutely. Yeah, across the board and it's a survival thing. You have to evolve, you have to change with the times or it's not going to work out. I was at Red Hat prior to launching Metafy and there's a saying, if you don't like change, you're going to hate extinction.

 

Will:

Sorry.

 

Mike Wagner:

You have to evolve, right? And so the traditional big box manufacturers, the proprietary game, there's a lot of value in those boxes. There's a lot of value in those servers. However, locking it in and keeping the software discovery provisioning lights out management, if you will, of those servers, that was always a big revenue stream and still is. So we recognize that and there is value there. Now, the issue comes in with the proliferation of white boxes and the desire just to have rather simple compute and the cheapness of that to be able to just chuck the server and pop in a new one. That's what really drove the dynamic change that's occurring now. And that's the technology commoditization curve eventually gets every proprietary special. player, if you will, right? It's just a matter of time before someone can do it a little less expensively. So how do you continue creating value around those platforms and around your brand? And I will say that HP and Dell and Supermicro and Lenovo and Cisco, they've done a great job doing that. These guys, they have big R&D arms groups, and they are always innovating and trying to make things better. And easier for their base. And so this is, it's not, you know, we're not trying to take over all hardware. This is, you know, just where it makes sense and where customers are wanting to have a heterogeneous environment and a simpler method to access that across different vendors. So yeah, it's been a fun ride so far.

 

Will:

I'll bet. How long have y'all been in business?

 

Mike Wagner:

Just three years now. Just three

 

Will:

Okay,

 

Mike Wagner:

years.

 

Will:

right on.

 

Mike Wagner:

Yeah. So that and, you know, so I guess to wrap up the enthusiasm from the players was, it's been from the start. So HP, Dell, and I think Lenovo or no, Super Micro, they were on the DMTF, the Redfish Forums board from the beginning. So it's not like the, you know, the board went after this by themselves. You know, they had the biggest guys on board with them. And you know, the communities that are associated. around Redfish are huge and they've got an annual user conference that's growing every year and all those things to make it happen. Same with OpenBMC, really strong following community growth. Everyone gets that the time has come to open these things up and allow customers and folks using the hardware to manage it as in the way that they want to, in the way that's most efficient for their organization.

 

Will:

Yeah, I think that's a great, I think this whole conversation is a great aspect of DevOps to bring up because it's one that doesn't get a lot of attention. You know, we talk a lot about cloud and Kubernetes and agile and stuff like that. But there is this whole bare metal component to everything. It's just abstracted away. But now we're also seeing companies, as we mentioned earlier in the episode, they're circling back around and leaving the cloud and going back to bare metal. So I think for anyone who's starting or interested in pursuing their DevOps career, this is definitely an area that should peak your interest and open up potential opportunities for you.

 

Mike Wagner:

Yeah, yeah, we've kind of, you know, from a programmer, so there's kind of the top down approach to, you know, people finding out about our tool that way where they're like, okay, so if this tool is in place, I can, we've got our back. So as a developer, you can be given the pool of servers, and you and a whole grouping of kickstarts or runtimes that you can load through our service catalog. And you know, it empowers you to do your job faster, right? If you need to tear down and rebuild different environments, you can do it a hell of a lot faster with a simple tool, instead of putting in a request to IT to have a server built that has XYZ and then six months later, you know, I mean, and that's some of the things God, you know, so, and that's the, so the funny thing about DevOps, it's DevOps is in the name, right? And

 

Will:

Right?

 

Mike Wagner:

it's the forgotten redheaded stepchild of DevOps operations. It's required to make the development happen. And that's where beyond all the infrastructure geeks that are actually working with the metal and working with the servers and having to somehow surface those and make them available to their developers. It can come from both sides. So obviously the value from an IT operations perspective is extremely evident, but it's also and we're working with Red Hat as an example and we're going to be going to a few Red Hat user group meetings to make it obvious to the developers as well that if it's a developer Could see the value in it just as much as the guy in operations because it does make things much faster much easier So we're putting the ops back in DevOps, making the ops

 

Will:

nice.

 

Mike Wagner:

easier.

 

Jonathan:

One of the very first guests I had on my Tiny DevOps podcast a couple of years ago by now, the topic was that basically that a lot of people forget to pay attention to some of that ops stuff. And my guest was specifically talking about, even if you're doing everything in the cloud, everything. You're never doing everything in the cloud or almost never. I mean, there's a router or a switch in your office

 

Mike Wagner:

Hehehe

 

Jonathan:

that probably needs to have the firmware updated every now and then or bad things are gonna happen.

 

Mike Wagner:

Yeah.

 

Jonathan:

And you know, there's so much, I mean, everything virtual is built on something physical

 

Mike Wagner:

Yes.

 

Jonathan:

as far as we know, maybe at some quantum level, maybe consciousness not, but as far as our hardware is concerned, everything

 

Mike Wagner:

Hahaha

 

Will:

Thank you.

 

Jonathan:

virtual is built on something physical. And we forget that. And

 

Mike Wagner:

Yeah.

 

Jonathan:

we just kind of assume that it's just magic and it works. And I don't need to worry about the BIOS update on my laptop. I don't need to worry about the firmware in my routers. I don't need to worry. And a lot of people even forget about just what base image does my Docker container built on top of? There's so much below what we deal with that we forget about. And that's really dangerous.

 

Mike Wagner:

It is massively dangerous and we've seen those hits happen. You know, it's one of those things where, well, as an example, you know, we had a financial services company that, you know, they were open about issues that they were having and they had to put off a BIOS update and it came back to bite them and cost them millions because they simply didn't have the bandwidth to get the people in the field, to take the servers down to do the BIOS update and cycle everything. So that's the other part of this. There's a de-risking value to bringing on a tool like ours, like Mojo Platform, that allows you to do the things you've been putting off and allows you to get a real standardized and transparent infrastructure snapshot, if you will, that is. updated in real time. So, you know, it just gives you that visibility and ease of use that's been lacking for years and years. So, so yeah, you're absolutely right. There's um, it's all it's just somebody else's computer. The question is, you know, how are they updating it? How are they doing the things that should be done that need to be done to keep it secure and to keep it from breaking? And those are the two big things that we love fixing. Now, the other really fun part of this, from an infrastructure perspective, so you picture, OK, there's a router somewhere. How do you access that? Or there is a group of servers somewhere, from an enterprise perspective. It's almost all hybrid, right? There's very few enterprises that are sort of in the Fortune 500 of any real size that are all in the cloud. And then there's the people that manage that develop, especially the servers, it's this very dangerous thing because from a knowledge transfer perspective, knowledge management, they hold the keys to the castle. They could do, if you get them upset, the danger is right there for them to do some incredible damage. So,

 

Jonathan:

Yeah.

 

Mike Wagner:

and that's the other part of de-risking that we do here. It really is a knowledge management platform in the sense that... It makes all of the day-to-day operational things very easy, understandable, and documented. So you can see exactly what it takes. And you know exactly what's in the field to do these BIOS updates, to know exactly what firmware and chip levels the different servers are on, routers, switches. So yeah, that's another big piece of this is de-risking from a knowledge management perspective. So everything isn't dependent on one guy that knows how all your servers are set up and how all the routers and how the networks are run, right? So one or two guys. So yeah, this is an important part of change management governance and de-risking, you know, having your own infrastructure again.

 

Will:

Because it never fails and I can say this, having been that guy, that guy is usually pretty grumpy and sometimes difficult

 

Mike Wagner:

Yeah.

 

Will:

to get info out of.

 

Jonathan:

Ha ha

 

Mike Wagner:

It

 

Jonathan:

ha.

 

Mike Wagner:

it yeah you are you are proceeding at your own risk and the company's risk right because you're right that guy you know with great power you know they can occasionally have great egos and you know and then to their um I guess in their corner for a minute I mean you know there's a lot that's put on those guys and sometimes it you know they get stretched to the edge and Getting a tool like this in place makes their life a hell of a lot easier. And then as a director of IT or an owner, and this is the thing, right? We work well with companies. We're installed in a footprint as small as five servers because we're very inexpensive. So, yeah. It just provides ease of mind and true resiliency. around knowledge management when it comes to making sure that if one guy goes on vacation or quits that you're not just dead in the water or

 

Will:

Yeah.

 

Mike Wagner:

scrambling like hell.

 

Will:

Yeah, because in many instances that person, no one else in the company realizes the burden that they're carrying.

 

Mike Wagner:

Mm hmm. Absolutely. Well, it's that what are they called? They hit by a bus scenario, right?

 

Will:

Right?

 

Mike Wagner:

If yeah, if that dude gets hit by a bus, what you know, what about tomorrow? You know, who's going to keep things working? Who knows what levels things are at? Yeah.

 

Will:

For

 

Mike Wagner:

And

 

Will:

sure,

 

Mike Wagner:

that's

 

Will:

yeah.

 

Mike Wagner:

the other part. From an auditing perspective, it's just saves months and months of field work and digging to... to assist with insurance, banks, HIPAA, health care. Across the board, there's just a number of business sectors that require very specific auditing to be put in place. And we have a reporting tool that's built into Mojo platform as well that makes it really, really simple. So kind of covered a lot of bases with the tool when Ian kind of dealt with it all. And And then with a ton of feedback too from these enterprise customers to build sort of the dream tool for them. What do they need? And the best part about being REST-based and sort of modern is that as the new requisites and we help guide the direction of that as well for the specifications get brought into the DMTF or OpenBMC. As soon as those new schemas are ratified and it becomes available for and published through the DMTF or even before sometimes, we'll do custom work, we can bring that into the tool and make it available to everybody. So it's a really cool thing because as far as refactoring the code, there's very little that has to happen. It's just pointing to a new schema within the API itself. So

 

Will:

Oh wow.

 

Mike Wagner:

it's... Yeah, yeah, so you can

 

Will:

That's

 

Mike Wagner:

just

 

Will:

super

 

Mike Wagner:

onboard.

 

Will:

cool.

 

Mike Wagner:

Yeah. On board, new functionality, onboard, new features really quickly. Um, yeah, it's kind of, it's, and then they're getting down into the DCIM space as well, like I said, around, you know, thermals and heat and everything else. So it's really just keeps on expanding. The use cases keep on expanding and then the power of open having all that R and D kind of being handled by your customers, by the communities, by all the people feeding into these projects. makes life as you can't compete with that. You can't compete with the literally billions of dollars essentially being poured in man hours and development from an R&D perspective, having those open standards in place.

 

Will:

This has just been eye opening for me. Like having done data center work in the past, now to have all of those things that I hated about working in a data center, just be wrapped up into, to a single tool to give you visibility into like the status of the hardware and what's the firmware level and just that whole visibility, this, this takes a huge burden off of the the consideration and the workload of thinking about or pursuing a bare metal infrastructure.

 

Mike Wagner:

Thanks, yeah. That's a goal. I mean, you know, no more having to go up and hit F5 and do all that low level madness and

 

Will:

Hahaha

 

Mike Wagner:

or F12, right? Get the BIOS going. Yeah, I know we got it all built into the tool and automated it. So yeah, it's a big step. It's exciting. And you know, the best part is when the customers see it and they recognize behind their own firewall, oh, wow. I see exactly what this is going to save us. That moment, it just never gets old because the reactions are just awesome to see. When it gets implemented and you see it in working for them and doing all the things that they wanted to do, it's just a very fulfilling, very rewarding part of having a startup for sure, having a software product.

 

Will:

Yeah, that's just mind-blowing. So you mentioned a couple of times your service catalog. I would imagine you have in your service catalog, here's the latest firmware and BIOS updates for these different things. But then you also mentioned OS management as well from a developer's perspective. So what does your software catalog offer at that higher level for users of the infrastructure?

 

Mike Wagner:

Yeah, so the service catalog has the more complex builds that would come after sort of a standard provisioning and runtime is put in place, if you will. So, and that's, you can do that either for us through templating, YAML, or you can do it via GUI. So we have both, right? You can do it either in sort of a mass gold image, if you will, gold build and... and really powered by Ansible for the most part. But we work with any of them. And then, I'm sorry, the question, the second part of that question was.

 

Will:

just like types of things that are in the service catalog, like

 

Mike Wagner:

Ah,

 

Will:

from a developer

 

Mike Wagner:

sorry.

 

Will:

perspective.

 

Mike Wagner:

Gotcha, yes. And then so that's the, so say, an OpenShift cluster on ESXi, something like that. So we've got that dropped in, built out in the catalog. Whatever your environment requires. So in virtualization, we're great. We work, we're a VMware partner. We work with Tanzu, ESXi. That's a very common build for us. But again, we're beneath that, so we're good with that. If a virtualized environment is what is optimal for your applications and for that particular use case, great. No problem. So yeah, it's the up stack associated builds themselves that we can add service catalog elements for.

 

Will:

Gotcha. And you said it's all primarily using Ansible. So literally anything you can build with Ansible can become a product in the service catalog.

 

Mike Wagner:

That's right. Yep.

 

Will:

Wow. Holy cow.

 

Mike Wagner:

Yeah man, fun times indeed.

 

Jonathan:

So I remember, here's a little anecdote from my time in the data

 

Will:

Back

 

Jonathan:

center.

 

Will:

in the day.

 

Jonathan:

Yeah, back in the day, I was so proud of myself that I went on eBay and bought a Port Master 2. Do either of you know what that is?

 

Mike Wagner:

Wow. Uh, no.

 

Jonathan:

So this was 2005 or so. Portmaster IIs were probably in their heyday in 1995.

 

Will:

Hahaha

 

Jonathan:

The Portmaster III, I remember using when I was at my first real job out of high school, it was a dial-up ISP. The Portmaster III was a Lucent piece of equipment, I think it was Lucent, or

 

Mike Wagner:

Mm-hmm.

 

Jonathan:

maybe it was bought by Lucent, but you'd plug a T1 into it and it would... provide dial-up access, 56k dial-up access on that provision T1. Portmaster 3 predated that, so it was an expandable box with I think up to 30 RS232 ports on it.

 

Will:

Oh

 

Mike Wagner:

Wow.

 

Will:

nice.

 

Jonathan:

Or you could get an ISDN module, but that's not what I cared about. So I bought this port master two, I think I had 10 ports, the one I bought on eBay. And that was my KVM switch for these

 

Will:

Shhh!

 

Jonathan:

servers sitting a four, four hour drive away.

 

Mike Wagner:

Wow.

 

Jonathan:

So I set up Linux on the serial console so I could do almost everything I needed.

 

Mike Wagner:

Yeah.

 

Jonathan:

With this now it wasn't really managed. It was just a serial console

 

Mike Wagner:

Wow.

 

Jonathan:

and some of our bios is a supported serial. consoles from the boot screen. That was the best. We had some old servers that didn't have that. So as long as Linux would boot to the kernel,

 

Mike Wagner:

Mm-hmm.

 

Jonathan:

we were golden. But whenever Linux failed to boot for some reason, corrupt a disk or a bad upgrade or whatever, we had to hop in the car, drive three to four hours to the data center and get that thing running again.

 

Mike Wagner:

Man, that is a, that's a setup. That is a, that's a hell of a way to do it. Yeah.

 

Jonathan:

We had somebody at the company trying to convince us to set up an IPKVM, which would admittedly be better because then you have direct access to the video signal and the keyboard. So anything you can do with a keyboard monitor, you can do. So it gives you a little bit more. But I was like, I don't remember, several hundred or thousand dollars at the time versus this $20 eBay purchase that gave us like 98.6% of the functionality.

 

Will:

at

 

Mike Wagner:

It was

 

Will:

9600

 

Mike Wagner:

funny as I dared.

 

Will:

Baud.

 

Jonathan:

Yeah.

 

Mike Wagner:

So what percentage

 

Jonathan:

Oh,

 

Mike Wagner:

of people?

 

Jonathan:

we had 115K band. I mean, it

 

Will:

Woo.

 

Jonathan:

was plugged into something fast, yeah.

 

Mike Wagner:

Screamer. So what percentage of people do you think, when you're saying KVM, I'm betting they're thinking, they're thinking virtualization.

 

Jonathan:

That's right.

 

Will:

Yeah, probably worth clarifying the definition of KVM.

 

Jonathan:

I meant keyboard

 

Mike Wagner:

keyboard

 

Jonathan:

video monitor

 

Mike Wagner:

video.

 

Will:

keyboard,

 

Jonathan:

switch.

 

Will:

video,

 

Jonathan:

I don't remember

 

Will:

mouse.

 

Jonathan:

if that's, yes, that's

 

Mike Wagner:

Key,

 

Jonathan:

what it was.

 

Mike Wagner:

yes, keyboard,

 

Jonathan:

Yeah.

 

Mike Wagner:

video, mouse, boom,

 

Jonathan:

So

 

Mike Wagner:

look at that.

 

Jonathan:

people use those, especially gamers will use those if you want to switch between two, you have a Mac and a Windows PC or something, you want to switch between the same monitor. That's what I'm talking about for anybody listening. I'm talking about that switch so you can have multiple computers hooked up to a single keyboard, monitor, and mouse. So I had these 10 serial ports connected to my 10 servers, whatever number of servers they were, and I could telnet into that port master too, and from there I could connect to any one of those, those 10 or so servers.

 

Mike Wagner:

Very cool. That's a very wicked setup for the day.

 

Jonathan:

I love that adjective for it.

 

Mike Wagner:

Yeah.

 

Will:

Thanks.

 

Mike Wagner:

Very cool.

 

Will:

Right on. So we're approaching an hour here. Is there anything else that we should talk about? We should talk about the name Mojo. That's a super cool name. Where did that come from?

 

Mike Wagner:

You know, so Medify is a portmanteau of metal simplified. And Mojo platform was, it just, it was like magic, you know? So when it came to, and this kind of goes back to sort of that Mojo magic thing, it just get your Mojo back, you know? Austin Powers can't help but

 

Will:

Right?

 

Mike Wagner:

be a man there. It just kind of goes into that, right? The magic of, doing metal very easily behind your own firewall. DIY, save a bunch of money, it's like magic. So yeah, Mojo is kind of an amalgamation of all those things, if you will.

 

Will:

When I asked that question, I was hoping that it was going to lead to an Austin Powers reference. So

 

Mike Wagner:

It

 

Will:

did

 

Mike Wagner:

did.

 

Jonathan:

Ha

 

Will:

not

 

Jonathan:

ha

 

Will:

disappoint.

 

Jonathan:

ha!

 

Mike Wagner:

Yeah, man. Come on, Austin Towers now. That guy needs to make a comeback. The world needs Austin.

 

Will:

Right? Yeah,

 

Jonathan:

Yeah

 

Will:

for

 

Jonathan:

baby,

 

Will:

sure.

 

Jonathan:

yeah.

 

Mike Wagner:

simpler times.

 

Will:

Cool, anything else that we should talk about or you'd like to share with us about Metafy or getting your Mojo back? Yeah.

 

Mike Wagner:

Yeah, I think that pretty much covers a lot of bases. It's an amazing tool. Like I said, if you have from 5 to 500,000, 5 million servers, 5 million, that's

 

Jonathan:

million.

 

Mike Wagner:

right,

 

Will:

You gotta do the pinky thing.

 

Mike Wagner:

servers,

 

Jonathan:

freaking laser beams attached to their heads.

 

Mike Wagner:

It can do

 

Will:

Right?

 

Mike Wagner:

an amazing thing and we can connect to those lasers and remotely shoot those for you. So,

 

Will:

Hahaha!

 

Mike Wagner:

you know, Got you covered on every front.

 

Will:

You can

 

Mike Wagner:

Yeah, check

 

Will:

update

 

Mike Wagner:

it out.

 

Will:

the firmware on the lasers mounted to your shark heads using the Mojo platform.

 

Mike Wagner:

That's right, that's right. Dr. Evil is actually a big customer, so

 

Will:

Nice.

 

Mike Wagner:

we appreciate him and his team.

 

Will:

Right on. So let's move on to picks. Jonathan, I'm putting you on the spot. Have you got a pick for us?

 

Jonathan:

I'm gonna do some shameless self-promotion today.

 

Will:

Nice.

 

Jonathan:

And I'm gonna pick my new podcast. We're about to do episode number 20. It's called Cup a Go. It's all about what's new this week in the Golang community. And we've got our first corporate sponsor recently, so that's an exciting milestone. So we're doing really well. We have swag if you want a cute little coffee cup or a sticker or a wireless charger with our. new logo on it, check it out. Go to cupago.dev. You can find all the episodes there and find us on social media. We have an active Slack channel on the gopher Slack too, so that's a fun place to hang out. So that's my pick for the week.

 

Will:

Right on, yeah, you've been cranking out the content there. I keep seeing new episodes coming out all the time, so good job on

 

Jonathan:

I

 

Will:

that.

 

Jonathan:

am. Yup.

 

Will:

All right, Mike, putting you on the spot next. What have you got for us for a pick?

 

Mike Wagner:

Wow. So, God. So I'm a physics geek, so I'll just, I'll give you one of many cool physics topics that I love. Check out John Stuart Bell. He was the guy who actually created Bell's Inequality, which helped prove non-local. influence, instant non-local influence, and sort of did manage to prove Einstein wrong about one thing. So if you want to see the guy who proved Einstein wrong about a big thing, namely general world,

 

Jonathan:

Spooky action

 

Mike Wagner:

I'm

 

Jonathan:

at

 

Mike Wagner:

sorry,

 

Jonathan:

a distance,

 

Mike Wagner:

relativity,

 

Jonathan:

right?

 

Mike Wagner:

and the idea that everything has to be, there has to be a local influence, he's the guy to check out John Stuart Bell. So he's... very cool physics guy who essentially he and the Nobel Prize in physics actually went to the guys that proved his inequality. So the three guys last year who actually received the Nobel Prize in physics. But yeah, John Stuart Bell and the guy's amazing and he didn't seem to get his props. Nobody knows who he is, but he's incredible. He was

 

Will:

Right

 

Mike Wagner:

incredible.

 

Will:

on. That's cool. I'll definitely check that out because I'm a big physics geek myself. Going all the way back to my first job out of high school, I joined the Navy and became a nuclear engineer. So just have always been fascinated by physics and all the different aspects of it.

 

Mike Wagner:

Cool.

 

Will:

So my pick, because you mentioned guitars earlier and you've got the guitars in the background, I'm going to show off my Schecter guitar here and

 

Mike Wagner:

Woo!

 

Will:

it's got the Sustaniac pickup on it. So if you play guitar and you've never had a Sustaniac pickup, it's been the most fun I've had in decades playing guitar. And so the Sustaniac will does kind of what the name implies. You can note ringing out forever so you can play like the blues and the like Zach Wild Ozzy Osbourne type stuff where they just hit those squealing

 

Mike Wagner:

Nice.

 

Will:

harmonics and just let them scream on forever and get that simulated feedback without having to crank your amp up to 11 although you're probably going to want to do that anyway. So

 

Mike Wagner:

Yeah.

 

Will:

yeah, Sustaniac pickups. It's my pick for the week.

 

Mike Wagner:

Awesome.

 

Jonathan:

I haven't had any pickups since I got married.

 

Mike Wagner:

Ah! Dada-ta!

 

Will:

Yeah.

 

Mike Wagner:

Simple crash.

 

Will:

Right? And that's a good thing.

 

Jonathan:

Are you gonna play it for us Will? I see you holding the guitar there.

 

Will:

We can, it's not plugged in, but put me on the spot here. We'll do a little slayer

 

Jonathan:

We

 

Will:

riff for

 

Jonathan:

weren't

 

Will:

you.

 

Jonathan:

prepared.

 

Mike Wagner:

Oh yeah?

 

Jonathan:

Awesome.

 

Mike Wagner:

Next time you gotta

 

Will:

Here you

 

Mike Wagner:

plug

 

Will:

go.

 

Mike Wagner:

that puppy in and give us some distortion to make that thing come

 

Jonathan:

That's

 

Mike Wagner:

alive

 

Jonathan:

right.

 

Mike Wagner:

with Custaniac happening.

 

Will:

Right? Because

 

Jonathan:

There we go.

 

Will:

that clearly did not do slayer any justice. But

 

Mike Wagner:

Hehehe

 

Will:

that's what we'll go with for this

 

Jonathan:

All right,

 

Will:

episode.

 

Jonathan:

we're holding you to it next time, Will.

 

Mike Wagner:

I could hear the

 

Will:

All

 

Jonathan:

We'll

 

Mike Wagner:

tight

 

Will:

right.

 

Jonathan:

hold you

 

Mike Wagner:

cording.

 

Jonathan:

to it for next time.

 

Mike Wagner:

It was fun. All right.

 

Will:

Cool. Thank you, everyone, for listening. And we will see you all in the next episode.

 

Mike Wagner:

Cool.

 

Jonathan:

Cheers.

 

Mike Wagner:

Thanks guys. Have a good one.