 You should be seeing our agendas. Everybody seeing our agendas? Yes. All right. So I'll officially start. And first of all, I'm going to thank everybody for the OKD marathon working group thing. We on the day of I haven't looked to see the long tail stuff and I'm still working on getting all the videos edited and uploaded. I think I've gotten three, one of which is a combo, the first one that had AWS and Charros, Eclipse Chey and it that just got up last night. And but on the day of we had over 10,000 views of the content, mostly coming from YouTube and Facebook, which is really always strange to me. But there is obviously an active open shift user group on Facebook and all over ACPAC and other places. So that's quite interesting. So that I think was really pretty awesome. So I wanted to thank everybody who did that and apologize as always for being late and tardy in getting the videos up. Day jobs get in the way. But we will get them. I'm working on the Dusty's one right now in the background, rendering it and then uploading it. So that was great. And tomorrow, if you don't know, we have a CNCF session going on that we I submitted and they accepted it was in. It's tomorrow at 7am Pacific, which is 10am Eastern, and it's called CNCF has 99 Kubernetes distributes. And this is why we built one more. And it's OKD and Fedora CoroS folks. Dusty's going to do the Fedora CoroS side and Charo and Christian and Vadim, I think is also you're going to get to do the demo. So I saw the slides from that. I did a few tweaks on it that you shared with me this morning. I'm not sure Dusty has inserted his content yet to the slide deck. Vadim, are we still waiting for him? I think Dusty's done it and I'm still working on mine actually. Did you guys share the right one with Diane? Because I think the I think the link that you would send earlier to Diane Vadim was. This is the one I was tweaking. So maybe I've got the wrong one CNCF. Yeah, that's the wrong one. All right, well, I'll stop tweaking it then and you can send me the right one. And that explains a few things because I got to the CoroS section and I was like, hmm, pretty empty right there. So, cool. All right. Well, I will await for that and and hopefully get some time this afternoon. I am going to the dentist in an hour to get a crown put in. I will be grumpy this afternoon and not really polite probably in my constructive criticism when the Novocaine wears off. So, so that's, that's good news. Once you have that slide deck done the other good news, at least to me, maybe not to everybody else. We did have a talk on the same topic, excepted to devconf us and which I submitted myself Christian and Antonio. I'm going to kill his last name. My dog, my dog she more. Thank you. Two points for you. I kill everybody's name. So I apologize in advance and it doesn't matter whether it's, you know, North American, European, ACPAC, LATAM. I just kill everybody's names. I never really been great at it, which is hilarious for someone in my role. Probably not. But so we have to record that again. So we'll do one for CNCF, which should get a lot of lift and awareness from the CNCF folks. And then devconf, I think they have a deadline of recording and uploading in the next two weeks as well. So practice and then we'll pull Antonio in and make him play the part of Dusty and do it again. And we can do that at our own pace. We just have to set up a blue jeans and record it. And then I can edit it and upload it for the devconf folks. So if those of you who don't know devconf.us 2020 is here. And it is a really fun place to hang out for a couple of days virtually if you haven't been, if you haven't been to one before, the one in the Czech Republic is one of my favorite conferences of all times. And it's not just because the beer is really good and really cheap. It's a lot of fun. So hopefully we can make the US version of it just as lively and fun as the Czech one has been for many years. So that's sort of what we have on the marketing publicity side of things. I know, Charo, you have been in onboarding sessions for the past couple of weeks. I think you've come up for air now. Yes, I have. Okay, so maybe next week we could take a look at adding the cookbook and recipes to at some point just set up a one on one with you. And I can show you how to update the OKD website and we can figure out what sort of recipes and how to incorporate also the videos that we've by then I should be done with them all onto the site and get that up there. Alright, motoring on. And thank you for your patience with my process today. OKD for update Vadim and Christian you're here. Where are we at? What's, what's the latest? Yeah, I can give an update. Most of this part of this two weeks, we've been fixing four to six nightlies. I think it's in four to six OVS is now running on host, which means upgrades will become much smoother. But we had to do quite a few tricks in OKD to catch up with what our course does. One of the problems we've hit is that network manager needed in Fedora needed a few fixes. So we requested those waiting for those to land and things should get back. The next feature would be samples operators finally using the community, meaning 10 to S, UBI 7 and UBI 8 image streams. So that would be different from ARCOS, and we will be able to contribute more. So while that's in the works, we're still testing quite a few things, see how that ends up. And the biggest problem last week has been that I was signing the I part has been broken for quite some time. We gave this and fixed it just recently. So this week we will be able to create a new stable release. So we didn't do that last week because we will hit the very same problems. Also, amount of changes in the latest table hasn't been that hasn't been anything critical. So we decided to slow that down, which leads me to the question for the community. That's a pretty extraordinary situation, which means we should somehow track it. The question is, should it be a calendar, for instance, or a GitHub issue? Or how folks should get notified about the table issues and stable problems? That's pretty much all I have to say. From my side, not a huge amount of news, but making some progress on the operator enhancement. I've actually joined a team now, the Windows machine config operator team, where I'll be working for one quarter. And with them I will pioneer the way how we release operators to OKD. So we want to make OKD kind of our first target for the Alpha version and make use of the different streams we have on operator hub. There's a few things, which is why the enhancement isn't up yet, with regards to releasing operators. We want to have a clear story that all the teams can follow. And the operator hub still uses the app registry for releasing and publishing operators, while Rated CoroS has already migrated to the new way of doing it, where you have a container that references all the images for each version, which is a bit more self-contained. I'm not sure how easy it will be to get all the Rated operators onto operator hub, with them already releasing the new way. So what I was going to do is reach out to the operator hub team and find out their timeline on that migration on that side. And then once we have all the things we have to do listed up nicely, there will be an enhancement that all the teams then can follow to publish their non-core operators to operator hub. So yeah, not a lot of visibility there right now, but there's definitely some progress. So can we go back to the last thing that Vadim said? You were asking for community input on how people wanted to be notified about an issue. Can you clarify what you're asking there, Vadim? Do you need people? Yeah, we have committed to releasing the stable snapshots every two weeks. But due to some issues we might want to deny that. The question is how the community should get notified and how it should be tracked. I'm thinking we could reuse the very same calendar we used for the meetings and file probably a GitHub issue. If folks have any other suggestions, shout it out. So let me see if I heard that correctly. So would we be using the Fedora project calendar to announce releases or delays and also logging an issue? So maybe combining the two. Is that the suggestion or are you talking about having another community page keep like this one that's in front of us now? Yeah, I'm thinking calendar events and the GitHub issue probably that would be a best combination. A dedicated mailing list suggested. I'm not sure how I feel about that. I don't feel like mailing. I'm not sure we need a dedicated list for it, but I think sending it out on our Google group list would make sense. Just to notify everybody who subscribed to that. Yeah, I think at the very least sending a note there. Maybe what I would suggest is that you do one, let's do one announcement using the calendar and logging an issue. And then we see how people and do to the mailing list to the Google group, which is basically a mailing list and see if that works. Okay, this sounds good. I think we cover all the cases mentioned. Yeah, so, you know, whatever. So, you know, there are other people have other ways they want to get notified. I mean, really, that's our channel to reach people and I hesitate to create any new channels or new mailing lists. I got enough of them. I'm sure you all do too. The only other thing I could say is in the Kubernetes Slack channel on the topic, if there's an urgent something or other, you might be able to set the topic. But even that seems like it's overlapping with too many other conversations that are going on with other flavors of OpenShift. So I think that's what we have is the calendar, the Google group and issues. So those are the three. Are you talking about opening a GitHub issue in OKD or in community? I'm thinking OKD would be best. Okay. Yeah. I don't think too many people follow the community page other than me to drive this meeting. So it's probably the safe place. So let's try it once and see if that works and if someone pipes up and says it didn't. There's a bunch of you on the call right now. If anybody has an opinion about that, let me know. All right. So I was just going to open this wish list again in the new tab. So did any of these get migrated over any of these on the wish list in the past little while? Venim or you said you were working on the with the Windows containers folks. We're working on a generic strategy how to release all of these operators and community, involve folks into the testing and try to avoid overloading the redhead because they don't like additional work. Nobody does. So wish list is useful, but we need a generic strategy how to approach it. And most of these items would be fixed once once we have a proper way of releasing it. Got that. So the other thing on the agenda, if we can move on, where where are we at in terms of outreach to the Fedora containers sake folks. I think Christian, if I recall correctly, you were going to do some outreach to someone on that side of the house. So yeah, I was actually going to use their work. They had like a meeting a workshop meeting. I was going to use that, but that was scheduled on a Saturday, I think two weeks ago. So yeah, I missed that and I haven't heard back from Climal yet again. I'll ping him again and try to set something up with him. And yeah, there's also one email where I think the operator hop folks were going to teach how to use their technology. So I'm not sure there's a few things there and we still have to find a date of that. I went to the meeting. I'm actually the only, I think person from here that did go. It was like two hours. Quite friendly. I think they would be overjoyed to have more people doing things. And they didn't seem to matter opinionated people as well. I took the Maria DB and updated it to run with the Radora 32 as the base. And that seems to be working, although I haven't yet figured out how to run their tests or really where to put it when I'm finished with it. You know, a lot of the processes aren't totally clear to me. But they released some some minutes out of that, which was attached to the open issue that everybody saw. And I didn't hear anything specific about a future meeting. How well attended was the meeting. There were four people including me. I think that's probably yeah, I'll definitely try to schedule it under the week on a work day next time. Not too early. That was kind of time. I mean, I don't particularly mind Saturday things if I, you know, if I know it early enough that I'm supposed to. I didn't know it happened until after it did. So that's why I wasn't there at least. Is there a fedora calendar for the meeting. They didn't put one up. So, like, that's how I didn't know. Usually, those things show up on the calendar and I wind up being subscribed to them and then I find out but that's not what happened this time. I think it was just kind of thrown together and then just everyone. I did the little chart thing he's saying what slots are good for me and then I don't think I ever heard from anybody about what time they picked. And then I find out like the next day that hey, this thing happened like what. And so that's that's kind of how that happened for me. So maybe also think about inviting Clement to this meeting next time in two weeks time. And then he can maybe pitch a little bit better, a little bit clearer what they need us to do. It sounds like there's some documentation just like we need documentation for the operators. They need some documentation like the MariaDB example that Bruce was talking about is like, what do you do with it once you've built it? And how do you get it, you know, where do they want to put all that kind of stuff. So that would be helpful. All right. I think we are already past the road to beyond we're beyond the road here. I haven't heard much more about okay D for IOT and it's sort of been suggested that that's probably not where we have resources to work on it. So unless there's a big community push for it. I think that's kind of an on hold topic for now. I'll reach back out to Maya and see where she's at and to Peter Robinson who I owe a bottle of maple syrup in exchange for some headache pills from the UK. So he we've been going back and forth with illegal trade between Canada and the UK during COVID. So I'll reach out and see if I can get them at least connected and see what their update is and make sure that Maya's got what she needs. But I think the K3S stuff that's going on in the CNCF is probably going to cover off what people need in that scenario. I don't know, Neil, if you have any insights into the if there's any work going on to merge Fedora for IOT and Fedora Coro s or any progress. Oh, I believe what's going on right now is that Fedora IOT is using the same kind of pipeline and set up that Fedora Coro s does but it is being managed somewhat independently. Because there is not. No, it's a it's a totally different pipeline a totally different build system. And actually Fedora IOT is they're using they're using what I don't think so. I'm pretty sure they're using closer to actually produce their image. The only difference between IOT and Fedora Coro s is the package set and the fact that they support Anaconda to install onto a system. They use something called OS build to create their artifact. Oh my God. At least they're they're moving towards that. And yeah, behind the scenes there is some some discussion right now how we can move COSA and OS build closer together. But it's not anything we can do in the short term. So yeah, the build system is different, which is why that is difficult to manage and OS build unfortunately doesn't really support. Anything the atomic manifests we've been using so far to declare our package sets and everything. So that's kind of anybody getting close house anybody getting OS build to work inside of the Fedora pipeline because bear. All right, you know, clearly nobody wants to tell me anything when I'm working on all these image build tools inside of Fedora great. So that makes number eight for Fedora image build tools. In production. Hey, what's another one. Right. Make it an even dozen. There's so many ways to build stuff these days it's crazy. You know, but it's got. So there's definitely some some plans on the open shift side to enable users to build their own images on their own clusters. And we want to kind of in doing that we want to bring OS we want to use OS build for that if we can. Right now, there's just too much of a difference to really get there quickly. But I hope because yeah, Cosa and OS build Cosa is CoroS assembler and OS build was built. They do the same things but Cosa is very much opinionated for building these CoroS ignition systems. And for example, the big difference between CoroS and Fedora IoT essentially is that in CoroS we run on the first boot. We run this. This init Rd. So from the RAM disk we kind of do the ignition. Whatever ignition does to create all the files it needs to do it does that in the init ramfs and then kind of lays that out and boots the the actual first boot the real first boot. Then goes into the desired state already while on Fedora CoroS we don't have this init ramfs run. So it actually does the ignition run in the real root on the first boot. Which is kind of hard to reconcile these two concepts and Fedora IoT does it I think for lack of of RAM. Yeah, so it's going to be interesting to see how that goes but it's it's definitely like on a long term roadmap to to merge these two projects somehow. All right, well there. That's a can of worms that I just opened so thank you. Let me close it. And this experiment topic here Vadim. Is this still a valid agenda item or is this something that we've an experiment you've done and. Yep, some of those are for grabs. The most. I'm easiest to start with would probably be C groups V2. It needs some messing around with. There are no arguments but in four to six we should all have all pieces in place. And it should work from after a few modifications. The rest is pretty complex. Okay, so it's good. I guess my question is, do you need me to leave it on the agenda and check in with you from time to time or. We don't have to have this in the agenda. It's it might. Later on spone up something interesting, but we don't have to check in on any of those every single meeting. I'm going to put it on the experiments list. All right, CRC for okay D4. I haven't heard any other other than what we talked about a couple of times in the. The day long marathon, I'm pretty sure this is still not available. Is that correct? I'm actually going to try this afternoon. Another build. Ironically enough. I've got a little bit of band width and so I'm going to I forked a copy of SNC and CRC again. So we'll see what happens. Okay, I will leave it on for another 2 weeks and then we will see if you know in time and if you get it, that would be great. Yeah, if we actually get it, if it builds an image that I can actually run with the CRC executable, then we'll have to figure out. Where we can host it so that people can download it and all of that. I have space on okay D.io. And, you know, maybe we can do that if you. We have this idea of a thing to download the where we put it and where we link to, you know, we can we can figure that out next week, but I would happily revive a link underneath here for for something. So yeah, finding the actual place where we should host it, which could be just in the okay D repo. Okay D for repo or the GitHub open shift slash okay D repo. It's about the other piece of the puzzle is you may get it running once, but maintaining it over time. Like I know from past experience with many shift initially landed on the team that I started out with that red hat on the dev evangelist team to maintain our mini shift, which was a real pain in the derriere over time. So the I think that the thing is let's get get it up, put it on there, but we also have to think about who who's going to maintain it and continue to build it for every release and every, you know, iteration. That's, and I think that's where we have to reach out like to Steve Spiker the PM and say hey, can you make this again part of your build process so it just happens. So anyways, but that's actually really good news. So thank you for that. I will leave you on the agenda for another week and go along here. News. It's going to be an attempt. It's okay. Attempts we like getting all clouds enabled with images any progress anyone know about getting Azure's Fedora Coro s images up there. I'm sure that's still just hanging out there like the laundry. That's still nothing. I think it's going to do. Yeah. I kind of want images like laundry list trash bucket, things like that for this agenda. The release timeline right here. The rest of these things I think are mostly things that documentation pieces bits and pieces. And I think when we start looking, Charles, I'm going to use you. We start looking at the cook cookbook and recipes to start seeing everything that we did in that. Okay. The marathon. It could be documented somewhere or if we do it as recipes, that's great. But also just to review and see if there's ways we can insert it into the documentation. Has anybody been doing any testing on okay D not on IBM cloud because we can't but on IBM Z and power systems. Is there anyone on the call from IBM. Would that be possible because I don't know if we do even have bills for those architectures. No, no, we don't. Yeah, so the answer to that would be definitively no. No, but so basically the question really is, is there anyone from IBM on this call right now. And we're not going to be getting okay D on IBM cloud the managed hosted thing anytime soon. Probably in my lifetime. So, yeah, I'll reach out to my contacts and still on my table then I didn't manage to get anyone to show up. All right, I'm just trying to clean up this list the last thing on the list replace ironic stub images with ironic built from RDO. Is this old and dead because I think we saw an open stack deployment work. That was that was actually needed for IPI on bare metal. Okay. Was it done. Yeah, that's definitely still an open issue. Okay. Yeah, I tried to find out what which images. I don't know what you're thinking about here but I'm not, I'm not sure maybe. Yeah, maybe Vadim, we can get together and discuss this, sometimes soon again, so we can back the team with releasing images, we can use. I think I tried to bug them in this issue. We can take care of that it's relatively easy we just need a lot of full requests. There's a lot of names on here so you know maybe maybe one of them will help us. Could you paste the link to that issue oh it's 197 okay perfect. I didn't see the issue. Awesome. Okay, so I'm going to close this window and go back. So that's that's what I have on the list here really of things I think that are open. I just would then like to open up the call if other people have issues or things that they've run into or if there's stuff that we should have put on the agenda that I didn't get. Where are other people at. I'm going to go around and I will stop sharing and see some faces and see the stuff in the notes now. Was there anything else that we should be talking about Vadim. Christian, Neil, Zed you want to introduce yourself since you're brand new to this group. I can do I mean I don't think I really have any qualifications to be here I joined the slack recently and got the little notification that this was happening so I'd come and say hello. But yeah I'm in the very early stages of learning about okay D and and what better way to learn than just jump in right at the defense. This is the place to be. Are you a red hat or learning are you from external. Oh no I'm external I work in online gaming and. All right games powered by open shift somehow. There you go. Well, we are considering it actually. That figures yeah. We'll do it on okay D you'll get better faster newer fresher everything. That's how games like to go. Gaming. Yeah, that brings up another idea. We passed on the heels of our live stream day, day long thing. We chatted a little bit about hosting an hour maybe monthly. For the okay D working group or maybe and plus fedora Coro s or other people. On our live streaming that's happening so. And so if you guys are familiar with twitch TV you gamers you. That's something that's an opportunity as well on me does take time. But I'm happy to to curate that and if we set up a schedule and people wanted to talk about their use cases for okay D or. You know some new deployment I know we still didn't do a Google cloud deployment yet and I think there were a couple others that we didn't get to. Like the Alibaba one and a few so we could. We could grab an hour on a monthly cadence if people are interested and willing to. To present what they're doing with okay D that would be lovely and it's just basically. It's like a talk show with live demos so. If you're if you guys are game for that I can we can use the fedora calendar and and try and get a slot. I know everybody's pretty busy and I keep tapping on Christian and Vadim and Charo and dusty. So I'm going to try not to bug them too much. I think this is something for other people who are part of the working group. If you want some airtime for the work that you're doing. Let me know and if I think if we can get like a commitment of four or five topics. Then I'll go to the live stream calendar and steal an hour and start booking it. Just ping me if you have a topic that you think and it's got to fill an hour. Because they don't like to do anything shorter than an hour. So. That would be that's my offer if you're interested let me know Bruce. Maria DB what you're doing at BC it anything like that. You know just you use it what you're doing at data not Neil what's going on in fedora that impacts OKD. Any or even the Clement from the container group might be a good guess to explain why they why it's important to join that. So there's there's lots of things and we can combine a couple people to fill an hour. People don't have enough content but that that's another opportunity to you know up level awareness of of OKD out there. That sounds interesting to anyone. Or is everybody's calendar so booked up that they're just not even thinking about it. It's interesting. Okay interesting is good. I'll just keep I'll keep reminding people and if you if you ping me send me an email at D Mueller. At red hat.com if you have a topic that you think you can you can use that hour for and once I have four or five. That's usually enough for me to be willing to. Bug the live stream calendar folks and get a slot for us. So are we ending short today we're you know we're almost. I've got a little something to mention to folks if they're interested please do go for it. Sure so. I apologize for dropping out for the past couple of weeks but my wife and I moved into a new house and we had our baby a couple of days later. So things are yeah thank you things have been a little bit crazy but I have been working on basically a series of scripts for. These fear. UPI that from start to finish get everything configured so the first script actually downloads. The F cost image creates an installation folder pulls down the open shift installer uncompresses gets everything in the. Like within this installation folder and then there's another script that generates the ignition files and then does the modifications on the manifests. And then I've got another series of scripts which I posted before but I've modified that. Go VC to bring up the F cost images in these fear and power them up and then from there you can just run the installer right now I'm working on it. Some health monitors in F5 so for folks that have that fives they can do health monitoring it's a recipe for health monitoring. And also I'm doing a little bit of a write up on squid to use for your respective proxies if you don't have access to F5. So all of this is going to go into a recipe of some kind. If anyone's interested in shipping in there were a couple other folks that were doing UPI on these here if you're interested just touch base with me. And my goal is to have something that's as easy as possible to get OK be installed UPI on these here with just running a handful of scripts. So let me know if you're interested. Jamie what can you type your email in in the chat there just for a sec. We have it there and I think that you know Charo and I are going to somehow find some time and start and set up that recipe page. And maybe once we set it up the structure we just figure out a way and base give people access to add their recipes to it. And so you could do that and then people can collaborate with you there that I think there's got to be a simple way of doing this all. But yeah, that would be great. Are you pull requests. Yeah, add recipes. Yeah, it shouldn't be too hard. Jamie are you running OKD and production at you mesh now. Not yet. So I ran into an issue with my f5 that I'm working with. So I was trying to build it the past couple weeks hoping to fix that f5 issue in the next couple of days and then we will be so it's going to be used. Basically as a educational tool for OpenShift for the University of Michigan community. When it's so sometime in the next week or two basically once all the bugs are out and it's configured. And there's some redundancy so that I'm not the only person at the University of Michigan who knows how to build this. And we get some backup people say it is another you mesh guy who's a CNCF ambassador I think I'm trying to think was a Bob I think is his first name. Do you know him. I don't know who that would be Bob. I can look I can go through who. Yeah, so yeah, there's another you mesh guy that's looking it up now. So the ultimate goal is we've got two OpenShift clusters on campus one being run by my college college of language science and arts and one at a unit called ITS. And then the idea is that the OKD is going to be sort of the teaching tool to familiarize folks with Kubernetes and OpenShift etc. I put the link to the Bobby's thing and Bob Killen. Oh Bob Killen sure OK awesome thank you. Yeah, so I was associating his his nickname there, but yeah he's he's a good person to sync up with to and it will be lovely to. I think between Bob's expertise and understanding and navigating this the CNCF landscape stuff and everything else that's going on if the two of you wanted to do one of those OKD working group things. That would be that would be awesome. That would be a great example of it. And Neil, I know in your role at Fedora plus data and what you're doing, you know, maybe combining that would. I know you like to talk so we could probably get an hour out of you easy everybody smiles on on the entire thing. Yes Bob. But so I think stuff like that would be good topics. And especially we probably should wait until we get the recipes up and the videos that I'm still working on up and then kick it off there. Because I think the. The ongoing the things that we're doing like getting the operators going and getting a cadence of the operators and just a part of the flow of every release so that for operators and talking about that and showing that off would be a good thing to do when that starts. Like when we when we finally have a catalog that's full. But, you know, things along that nature will help drive people to the working group. And the hardest thing for me about OKD is it's there's no firewall. There's no lead gen. We don't know who's using OKD really out there in the wild. A few people have turned on the telemetry when they've installed it in production. Once in a while someone shows up on the on the dashboard and I go that's that's a win for me, but not very often. I think there's a couple of AWS folks who are running OKD, but they're all anonymous so I don't know. So if you hear of people using OKD out in the wild, it would be wonderful to get them to share their use cases and talk about it. We have now over 2000 OCP companies using OCP out there that we know about what we really just don't know. And pretty much we're not supposed to know because it's open source that people are supposed to get to use it anonymously any way they want. But it would be really nice to have a few more public references and use cases. And to get them involved in sort of end user stuff. So that would be great. Thanks, Ed. So that's our time today. I'm not trying to fill up time. I can give you guys 10 minutes back if everybody wants their 10 minutes back unless someone has some awesome demo they want to show me or in everybody else. Otherwise, you get 10 minutes back. And I am three videos of the working group meetings behind now officially. And someday I'm going to outsource all my video editing, but it's not happening yet. So you will see them go up on the OKD playing list eventually. I can usually get about one or two a day done. And now I'm done behind by about 12. So we're slowly working through them. The next one up, as I said at the beginning of the meeting, I should have Dusty's demo of Digital Ocean and is talking about Fedora CoreOS up later today. So look for that. Tweet it out. And as well, join us for the CNCF session tomorrow. It's early my time, 7 a.m. So let me just see if I can find the link. I think I put it on my Twitter channel here. One of us will have the link to it. Latin America one. And this will have the details somewhere. Must be under OpenShift Commons. But it would be lovely if some of you could join and listen in. This is the one. I'll throw the link to this in the chat tomorrow morning. And that should be loads of fun. I'll stop sharing, throw the link in chat. So you have the details. Feel free to retweet that on your social channels and get more people to attend. That'll make the CNCF happy. And maybe they'll give us a repeat visit too. So we'll see. And those are kind of hard to come by slots. So that'll be good. Anybody, the metal three cubed team. Anybody connected with them yet? Metal three cube. Anyways, the bare metal deployment stuff. Not yet blank looks. All right. I'll go grab someone of them and make them come next year. Next time. Yeah, that's actually where I started playing with it, but ran into that roadblock with the ironic images. And that's when I switched my bare metal to UPI. Yeah. So the metal three folks are trying to submit to the CNCF. And so, yeah, maybe if we can grab one of them, we can get that ironic thing figured out and do something jointly. Okay. D and metal three. And then you can leave that with your heavy metal band and your guitars there. Okay. So that was the other thing. So I'll reach out to them. They reached out to me. So I'll make them owe me something. Okay. It could be those ironic images. So we'll see what we can do. We need some webinar. Your tweet link. Who's asking the tweet you linked requires recreating account. Apparently. Yeah, you do have to register for a CNCF webinar. I, you know, for an open source foundation, they still have this fetish for lead generation and tracking people. And I did say that on recording. It just boggles my mind. But yeah, shakes her head and goes. Yeah, there's no, that's, that's CNCF for you. They want to know who you are so they can love you. And yeah, that's, that's, yeah, that's another story. And I don't want to go ranting on that today. Another day, Neil, you and I can chat about that. We love the CNCF here. We do. We seem to throw through a lot of them. I mean, there, there are really huge positives for having CNCF and having that umbrella organization and, and keeping things connected. But there are also, it's, I, I hesitated. I don't think it's the CNCF's problem. I think it's a Linux foundation process. Eager's problem way they think about community and I'll leave it at that. Yeah. I have, I have personally much to say about that if I ever had, if I ever, anyone ever wanted to know. Yeah, someday I'll have a podcast where I can talk openly about these things. Till then, not. All right, guys, I'm going to let you all go. I'm going to go get the dentist to drill my face off with it. So if you talk to me tomorrow and I'm grumpy, just bear with me. I should be fine. All right, take care all and congratulations on the new baby there, Jamie. And yeah, well, well, well done. A new GA release. Take care guys. Bye bye. Bye bye. Thank you everybody. Bye bye.