 The OKD working group documentation subgroup meeting for April 19th of 2022 and let's do a quick review of current projects of removing the community. Brian. Yeah, I think we're just trying to work out what we're doing in terms of community officials. That's the only bit I think is outstanding. We've agreed we're going to drop membership in terms of we don't really have an active membership role at the minute for each of the things. It's whoever turns up and wants to participate does. However, we do need to list the officials and I think there is the how do we get the officials and does every working group have a set of officials? Question, but other than those two, the community report can go. The two documents in terms of code of conduct and the charter that's in the OKD site now. So that report could effectively go now. So do you need someone to make a pull request to remove it or? We need somebody with admin access that can archive it or delete it, depending on what you want. How you want to do it? Let's archive it. So Diane, you have access on that. So if you could archive it when you get the signal, that would be awesome. Which then does leave us with the how are we going to resolve the office's working group? Let's send something out in regards to that to the working group. But I still don't have access to social media to be able to promote the fact that members will be able to vote. I can post something on the mailing list. We also want to post something on social media. So I don't have access to the Twitter yet and Diane, you were gone. But the just is what we want to do is have an official election of the, or not election per se, but yeah, I guess it would be election asking the chairs of the main group who they would like as the chairs for the subgroups, right? And people have thrown their hats in. Sandra has thrown his hat in for the virtualization. I'm assuming that CRC would be, I don't know that we have one for that one. But once we get those chairs for the subgroups, then we'd be able to put that into the document and close things down to the community repo. Cool. So I apologize. I thought I had already given you the Twitter username password. I guess it didn't happen. So I thought that happened like a month ago, which is about how long it's been since I've been on one of these calls. Do you remember how you shared it out to me? I don't. Boy, that would be asking a lot. Let me just see if I can. We'll deal with it async, but I'm trying to. Okay. Yeah. So yeah, so at this point then we can move forward. Just send something out that says these folks are looking to be chairs of their respective working groups. And we could do a binding, non-binding vote, kind of like the Kubernetes or the CNCF, a TOC does. So basically, people can voice their opinion, even if they're not chairs of the main group, right? Mm-hmm. That'll work fine. Totally sounds much more legit than what we've done in the past. So I'm happy to have it be legitimized. Good deal. Moving on to the next thing is technical documentation. How is that coming along, Brian, with your fork? So I'm still working on it. I have not done much. Sorry, John. I haven't accepted your full request. And that's just me not having time. What I'm trying to do is I'm trying to create a work example in the documentation. And what I'm looking to do is take the console module and just change the documentation link from docs.okd.io to okd.io. So we've got something that's very visible. It's obvious whether the change is in or not. And I'm going to do the two routes, updating existing cluster and create a new installer, which are the two routes that you're getting a custom built module live. So that's what I'm trying to do. And I just want to get that working. And then I'll explain how you do it. And then a worked example of doing it, which might work out to be a YouTube video. I'm not sure yet, but in terms of, but I just think there's too much information just to have it purely written as a sort of an abstract. So I want to do a worked example. And I thought something visual on the console is something that's very, very easy to understand. It's obvious if it's worked or not. Because if you hit the link and you get to the docs, it hasn't worked. If you get to okd.io, it had worked. I haven't got it working yet. That says most of the second way to do that. Yeah, I agree. Many because it was Easter and I had other things to do over the last weekend. So yeah, I'll work on trying to get that done. When I got it done, I'll push it all up, pull the, take the pull request in and hopefully we can then push that onto okd.io as the start of the technical documentation on the main site. So that's the plan. It's moving, but Easter break. I was going to say, unless anybody else has a suggestion of that's a stupid idea, do something different. To me, visual is always good. Yeah, that's a starting point. I mean, that way you have something to show and then you have something to build on. So then we can throw other examples. The installer itself is interesting to play with that. It just takes a long time to build. Yeah, I think we need a special thing on the machine config operator because obviously that's difference in okd with Fedora. I'm also interested in, I think an interesting one might be is the catalogs because when we get the okd catalog, which I'm hoping we're going to see movement on now. I think that'll be another interesting one in how do I add my own custom operator into a catalogue which I think then comes on to what we're going to be talking about a little bit later on SEF because I don't believe there's an easy way to install SEF to actually test it at the minute. So yeah, and then as we talked about in the last meeting, I need to get on and start collecting, debugging, testing, things, best practices, tips, tricks, hints, voodoo magic, whatever works to actually get it close to up and running once it's not running. All right, any questions or comments or further thoughts on that? I have one question about what we talked about just previously, removing the community repo. What is the trigger for that? Are we ready for that to happen or do you need more time? No, go. Do it now. Yeah, you've got everything copied over, so it's fine. Whenever we're going to archive it, we can always get it, get the stuff, we're not going to lose it. Sure. So archive it rather than delete it, then it's there, but it's closed for change. I just reached out to Christian Glombach has admin privs and he can archive it. Excellent. So archive, and I just requested. And working group repo, that's going to wait until things are removed. So I have a question for the group. If we look at the website, moving on to an item that I have, if we look at the website, if you look at the working group menu, it's like working groups. It's like, okay, do you working groups? Okay, do you working group? What I would like to do is you'll notice I added minutes in there. And actually, I get I need to fix the way I did it, but I started to add the minutes in there. I'm setting up automation for it. Because although we can automate from HackMD, an individual file going over, we still need to automate the updating of the index, right, so that people would have an index. Would anyone be opposed to me? And Brian, I'm looking at you in particular, changing this to the top menu, being OKD working group, and then a subgroups sub item that then links to the subgroups? Yeah, I think that would make sense. Because if we get more subgroups, then it gets, I mean, I was going to call that OKD working group, the main working group. But I think it makes sense to have a subgroups heading, and then all the subgroups underneath that. Anyone else have thoughts on that? Again, this is, if you go to the website, okd.io, it's one, two, five, six, seven items down on the left side there. So just changing that to be OKD working group, and then about charter minutes, and then subgroups. And then subgroups would have the subgroups there. Oh, so are you going to move about charter minutes into the working group? So move it all up, basically move everything up, and then have a subgroups submenu under OKD working group. Yeah, makes sense. Yeah. Or it could, we could even, it seems kind of redundant to call it OKD working group, we could just call it working group, the top menu item, because I mean, what other working group would it be that we would be hosting? Yeah, I look at it as a little eye weird. Right. Does that make sense, Brian? Yeah. I mean, in the thing I called it the primary working group, because I don't think we've got a name for like the main one. I mean, we call it the main or the primary or the, all the subgroups have a name, but we just call the working group, or I don't know whether we need something to signify it's the main or the primary one. I know. What do folks think? Because otherwise you've got another context, if I just say the working group, if I'm within the documentation working group, I obviously mean the documentation working group. So then what do I call the top level group if we want to defer a decision to, I think we use the word main group in our meetings. So I just think we have to sort of call it something other than just working group, because Diane, do you have any thoughts? You've been around these waters a lot. I apologize. I was trying to close that archive that thing, so I missed that conversation. So the idea is we always have the top level sort of working group and then the subgroups. What would we call the top level working group as a reference? We always say the main group in conversation, but if we're going to talk about it. I always think of it as the working group and everything else as subworking groups. Subworking group, okay. So that's how I've always referred to it, and I've seen other things done. We don't have a technical oversight committee like the CNCF has or the Kubernetes things. We're not that complex yet. So yeah, I would stick to that. It's sort of the dev working group too, right? It's sort of the overall dev working group. The subworking groups are looking at particular facets of things, not OKD as a whole. Yeah. I always refer to it in that when I post the videos online for this group, I always call it the subgroup documentation subgroup. And then the other one's just the working group. Okay. So then we can put a task on me to go through all the subgroups and change the name because if you go to the documentation overview, that says the documentation working group is responsible for that needs to be the documentation subgroup. Yeah. Subworking? I hate this when it's... Yeah, I say subgroup because there's too much working in it. I think it helps delineate them a little bit more if you just say subgroup within the OKD working. Yeah, that's fine. I'll do that. Okay. And is that... I mean, and you can make the change too, Brian, if you want while you're in there in terms of that menu, but I think that that will clarify things a lot. Yeah. In minutes, I accidentally did redundantly. It's actually like slash minutes, slash minutes, I need to fix that. I was doing this with a baby in arms, so I made some mistakes. Okay. There is a gotcha you might come against with the automation. Currently, when there's any change, it does a full spell check and it also does a link checker. If a site is down, like when Git was having problems, it blocks the publish. Right. So that's just a gotcha to be aware of. If you look at the actions, if there's a failed action, you may encounter that with automation that we... If you find some minutes don't get published. Yeah. Just be aware that that... We should put this in the little notes section of the... How to... I think there's like the how to contribute page that you set up or was it? There is a... But I'm guessing that's... Content guidelines page. We could put something in the content guidelines page. Yeah. Yeah. Okay. All right. Let's see. Next on the agenda is... Okay. So here's the things that we need to communicate out, right? So talking to Christian earlier today, we can't release any 4.9 since 4.10 was out. OkD has always followed this pattern of when that shift gets made and all of the infrastructure for testing gets... When it gets put in place as the main release, no more previous releases are done, previous minor versions are done. So a 4.9 release is not going to be forthcoming to deal with that SCD issue. The SCD... What is it? 5.0.1 to like 5.0.5, I think it is. So basically SCD under high load can drop data from some of the members. We've gotten questions about that. And so we need to communicate out that 4.9 is not going to be updated anymore. There won't be any future releases for it. And in the same breath, we have to point out that 4.10 is currently blocked by that bug there that I've linked to the bugzilla for. So Christian actually tried to do a release this weekend but ran into a 4.10 release this weekend but ran into that OVS issue. So we haven't really communicated anything out about how our release schedule has changed. And I think we should do that right in the near future. Because right now there hasn't been a release for almost a month. And people don't know that we're sort of changing temporarily or whatever, changing hands. So I'm thinking we write something up that highlights these two particular things. Do we actually have an OVS issue? I'm sorry, what? What was the OVS issue? So there's a link to the bug but it's basically an issue report binding. And so, so he just put in a bug on this and apparently this is 4.10. So there's a link to it in the notes. And Christian said that that's a showstopper for OKD release. All right. I reached out to him because I have a question about that. But okay. I know there's other issues I think that are out there that are stopping us going forward also. But okay. Right. And so that one. Right. And so, well that was his response in terms of why he didn't cut an OKD this last week. I wonder if he can delete the one that's out there. Because if you look on the release pages, there's a failed release on that page. On the release page? Okay. Yeah. I don't know if he can delete that or not. It shows everything failed but it's still out there. It's not the 3.7 and then the one from the weekend. Okay. All right. Let's add that to our list of items to do. So what do folks think about communicating out the changes in the scheduling? Oh, the answer was a Brian that you asked. It was, Vadim was cutting them so regularly like almost every weekend that I think people are starting to notice that there's nothing coming out. So it's inferred. It's not really. Yeah. Diane, go ahead. So yeah, it was inferred and now that he's taking sort of a bit of a break from that, we're going to have probably a expectation management to do. Do on the the okayd.io site, I think we made a provision for a bit of a blog. Is that correct Brian? So if one or both of you, Brian or Jamie want to write something up, we could put it there on the blog and then I could amplify it through the OpenShift and the OpenShift Commons Twitter handles and the and you could do the okayd.io one as well and we could get the word out that way. If you wanted to do a to finally use that blog for something, I think that might be a this might be an appropriate time to do so. That would be my my gut and then start. Okay, Bruce, I saw your hand up and then Brian. Yeah, no, I think it's not to my mind. The issue is not that it's been a long time between upgrades, but basically we don't have any 4.10 upgrades that upgrade from anything. If you look at the 371 and the reason that I didn't try it out is because the upgrades from everything fail fairly spectacularly and consistently. And of course, it also doesn't upgrade to anything. So it seems like not something that any sane person would want to try. That's certainly not in production. And you say that they upgraded all my all my clusters to 4.10. Well, I'm going by what the documentation says. Yeah, yeah. Okay, yeah, if you had good results, including Ceph. Because I'm well, that's an interesting one. Right, because I'm running that not for the registry like you are, but for provisioning, auto provisioning storage, I'm running Ceph. And so that would be a blocker for me for my production one. Bruce, can you do an F cost standalone test of mounting to your Ceph for us? Because Sree, John and I had asked as well. We were asking Sree to try it in that ticket and he hasn't responded yet. So you might have other things going on. Well, my Ceph is on the, like it's running on the cluster itself. Right. It's not standalone. Yeah, we need, yeah, Diane, go ahead. And it would take more resources than I have to create a standalone one. Got it, Diane. I just have a question for John Fortin. You did upgrade to, or you moved to 410. Did I hear you say that? Yeah, everything's on 410. And is it running successfully? Or is it? Yeah, except for Ceph, I didn't have any issues. The upgrade. How did you do the upgrade? I clicked on upgrade to 410. Okay. And I'm running beef for IPI. I didn't have any issues. It was clean as any update I've done and cleaner than most. Okay. That's the 3-7 one, right? Diane, you broke up a little bit there. Yeah, I'm going to turn my video off. I'm at my mother's house, so I'm using low-fi. Not even Wi-Fi. I would just go low-fi. There you go. So, and my nephew's here, and he's probably gaming downstairs. So, what I would suggest is that maybe we write a blog post talking about the thing, and then maybe have John Fortin, if you can add a few words about your upgrade process and, you know, like that. Honestly, if I knew now what I knew then, I wouldn't have upgraded. Yeah, maybe say that too. I mean, I'm never against saying the truth. That is, you know, the general surgeon's warning section, too. It's like, yeah, I did this, it went smooth, except now that I'm here, this is what I would, you know, warn people about. So, you know, that's, and I, you know, and that I think would be, if we can come up with a blog post, you know, over the next week or so, I'd be thrilled to have. And as always, honesty is the right way to go with this. You know, it gets you much further than trying to make it look like an awesome, everything went smashingly, and now I'm happy I'm here, party. So, yeah, so maybe that would be something we could collect with, even if we did it in HackMD as a blog post collaboration or something. Okay, so should we break these up into a separate document for upgrading to 410 and sort of where things stand with that? A separate thing that talks about 4.9, because that should be independent, because whether it's 4.10 or 11, 4.12, there should be something out there in the blogosphere that says 4.9 is no longer being updated. Then, since we don't have anyone that can do any SEF testing, should we send something out, even if it's not a blog post, in some fashion saying, hey, is there someone in the community that can test this for us? And what we need specifically is someone to try testing F cost directly mounting SEF to rule out that this is an F cost issue on its own. Well, I thought that Street City had done that. I saw you ask, did he respond? I saw you ask him, but I didn't. Let's look at that real quick. I'd have to go back and find it. Hold on, I got it, I got it, I got it. Yeah, so you asked him, and I had actually asked him, I think earlier in the conversation, or elsewhere maybe, of, I think you responded, yeah, so you responded two days ago. Can you try mounting the SEFSS mount in a container in a standalone F cost note? Yeah, that, yeah, he did it. I think he did a SEFSS mount directly on, directly on F cost. I don't think he did it in a container, and I'm wondering whether there's an issue with the new Creo and with SE Linux. All right, so I had that. So that would be the, I don't mind the next test. Okay. Well, that you could do, Bruce, because it's just mounting it in the container. Okay, so in standalone, okay, so you're talking about 4.10 though, right? Yes. Okay, so I can upgrade my test cluster to 4.10, assuming it works, no problem with that. Trying to think of visibility. He said there's a flag you could change that will make it visible outside the cluster. Right. Jamie, can you throw that discussion link in the chat? Yeah. I don't have a handy. And we don't need to get too deep into it here for Doc's meeting, but just if we, if we feel we can, we need to reach out, let's reach out to the community and see if we can get someone to test this. Okay, so what Jamie put up was just the thread link, which I have flagged in my. So things stopped right there with basically like, okay, can this be mounted in the container? And does it still have the issues? Saying that working was true. Yeah. We'll take a look and let us know. We don't have to decide that right now, but if we can't get anyone within the group to test it, let's, does it make sense? Let me ask, does it make sense to reach out to the community to try to get someone to test it? Because this has been lingering a bit now. Right. Yeah, the issue with that with the suggestion is the existing PVC mounts. And anyway, so I'll have to look at it a little bit more because I've got sort of a handful of existing PVC mounts, so I would hate to lose. You might have to look at creating a, creating a clean EFCOs server, whatever. That's not, it doesn't have, okay, de-installed on it and see if you can do a remote Creo mount on it or whatever. I'm not sure exactly how to do that. Yeah, I was just, I think if you're going out to the community, we have to be very, very clear what the ask is to make sure there's no ambiguous instructions in terms of, this is what we need you to do, A, B, C, D, E. Right. All right. Well, let's look at that and see what's, we can talk async on that one. Another one that came up is CRC, put ready containers. Are we, it seems like it's sort of stalled locally. I haven't had the time to deal with it. There's still mention of CRC on the, in the OKD documentation. And Charo did do a build in like November that apparently does not work anymore. Do we want to do a request for someone in the community to do put ready container build? Or does someone here or someone else in the main group have time to do one? Again, I think this is one of these because is somebody outside of Red Hat able to do that? Yes, they are. Well, because I think whenever there's been a change, the source and things like that. I remember the last time we build, he had to go and tweak something because there was a change in padora or there was a change somewhere along the line. So it may not just be a case of running the build script. And what are we building it on, 4.9 or 4.10? Yeah. Well, but also like all these tweaks shouldn't be necessary since there is a single node OCP capability, isn't there? I haven't tried it, but I haven't tried it either. I think it was originally like in 4.8 maybe and slipped 4.9 and presumably it actually made it in by now. In Google chat, there's a little thing where you can put your hand up. And I can't find it here anymore in Blue Gene. So I apologize. I just keep putting my hand up. So we're talking about CRC here, right? And I think the conversation that I had, and this is over a month old with his name, Charo. Charo, thank you. Brain is not complete. Was that people really weren't using it anymore, CRC? Yeah. So the thing is, as well, you were gone. We had like three different people pop up that were like, hey, this is in your documentation. Hey, this doesn't work. Hey, granted, it's only three people. But one thing we learned in, at least for me in radio, we learned it in radio. It's for every person that calls into the radio station. There's like 100 people listening. And I think you can use the same thing for anything, right? Okay. So I just wanted to say that out loud, because the last conversation I had, it was let's move everybody over to snow or whatever the acronym is for the single node. Right. And that. Yeah, go ahead, Brian. Okay. I was just going to say, one of the things is to be cognizant on listening to the what's next in OpenShift, they did mention CRC and streamlining it somewhat. Because I think the difference is, Microsoft is Microsoft, no UI. It is very, very much cut down. CRC has a lot of the stuff removed, like the machine set and things like that. So it's meant to be a lighter weight cluster. SNO is a full cluster. So you need much more resources, but it's on a single node. And then you've got the full node. So there you're four flavors going from full to micro. So I mean, when I started this, I wanted CRC to work. It's just that I didn't have a single laptop that was beefy enough to run CRC, because you can get it to run on a 16 gig MacBook Pro, but you can't do anything with it. And I also found that the file system I filled up and the Mac virtualization, the hypervert engine, doesn't allow you to expand the disk very easily. So you end up being bound with, you don't have enough resources in CRC to actually do anything useful other than start a server that you can use the console on and do nothing with. But I think there are quite a lot of people, and this was a discussion that come into the What's Next for OpenShift video. There are people that want a local cluster for testing and playing with. Again, a lot of people, they want to learn, they want the admin access, but within their work situation or their university situation, they get a developer or a user. So I think there's a lot of people who do want that local access, but they don't have a workstation with 256 gig of memory. Yeah. We've heard this story before. So if three people raised their hands, Jamie, that makes me happy. Three people raised their hands. So I just didn't want us to be having this conversation. If there wasn't any interest in CRC anymore. There were three people that came into the channel, and I think there was one discussion item asking about it. All in a short period of time, it was actually kind of quirky. It was all like within the same week. You take your toys away, you take a toy away and or threaten to take a toy away, then you find out who's actually playing with the toy. Yeah, this is true. Well, the issue that they had was that charros build stopped working, I think, because of a certificate issue. Right. Yeah. So Bruce, you had your word. It looked like it. Yeah. It will time out. I think it works for, I don't know, a couple of months, and then certificate times out. And that's been an ongoing, well-known issue, which means that the only way that it will continue to work is if somebody continually rebuilds it. But I think that one difficulty is that what you don't know is these people that are unhappy because they can install it, whether or not once they install it and see how it works with the resources they have, if they would then be happy. That's true. That's true. Like what I spent, I guess, the last month, I've got a module on Kubernetes implementation for my students. And so I've played with the MiniCube, which I guess all of the Red Hat demos use. And this time I use kind, Kubernetes and Docker. Both of those are very similar and work reasonably well. The MiniCube does have a console that you can install separately. And like I have once built the Charles thing, and it never worked well on my MacBook Pro with 32 gig of memory. And the, well, when Microshift came up, I tried that and it didn't quite work on the Mac. It's the DNS field to come up, but it was close. And it could be that if you add in the console as a separate add-on that you might get closer than to what the users actually want to play with. Like I think it's not just a matter of rebuilding something that exists. If we're going to take it seriously, we probably need to get a working group to slim it down considerably. Well, so I guess for the purposes of this group, I think we should figure out, right now we have documentation to something that doesn't work, and that documentation actually points to a Fedora repo that, I think the dado guys hooked that up so that we could put CRC up there. But like I think it was Neil, but Charo and Neil I think worked that out separately, and those details are not written anywhere. So it's not like we have access to go and delete that. Neil was the one that brought up Microshift too, right? Right, exactly, right. So I think what we, first off in the bigger picture, I think if we're going to manage projects or sub-projects, we need to have internal documentation about who has access to what and where things live, so that we don't get into a situation like this where we're kind of like, okay, how do we delete it? Well, we got to reach out to one of these other folks. And Charo's really busy right now. Neil will probably be at the main meeting next week. We can touch base with him. Charo has documentation on how to build CRC. He does, and I've linked that to people, and we could include that if we were going to try to reach out to people. But I guess my question is, do we want to try to reach out to the community to see if someone wants to help towards it? Diane, or sorry, Brian, then Diane. Yeah, I was just going to say the link to the Charo's things actually in the CRC subgroup section of OKD.io, I've linked to his content there. But I think one of the problems is the CRC subgroup never actually got established. We said we needed one, and I put it on the site. And I don't think there's ever been a meeting of it, and there's nobody really driving that. So I think that is one thing that if you're putting a communication out, we need volunteers to actually go love CRC. Because I think there is still life in it. It looks like Red Hat have plans for CRC, for OpenShift going forward in terms of making it more usable. So I think it's not time to sunset it, because it's an active project within Red Hat for OpenShift that they want, as Bruce said, they want to slim down. They want to make it more usable on a workstation. I think the positioning that MicroShift is an edge, it's not your sort of cut-down developer workstation. It's an edge sort of package. So I'm going to add on to that. I think I agree with everything Brian was saying is pretty much what I was going to say. I don't want to sunset it yet, especially if three people pop their heads up. And it is part of the conversation in the OpenShift roadmap. So there are Red Haters working on it. There is another group, and I'm just going to ask this question naively. Over the month I was gone, did anyone from the Operate First team reach out to Carsten Wade or any of those who didn't? Okay, so there is a group of people at Red Hat who are spinning up a community of build and CI CD patterns called Operate First, and they have a cloud that they can build stuff on. I told them that when I was well and not traveling that I would get them here, because I was hoping that with the problem of like Vadim and everybody else to be able to do community builds of things that we owned and managed, I'm thinking out of the box here that maybe their cloud is based off of mass open cloud in Boston. It is an OpenShift cloud. They have resources and they were looking for projects that maybe getting them to take Charo's information might be a good candidate to see if they are viable partners to collaborate with. They'll give some resources and if they have interns and stuff at Boston College or whatever that would be interested in this. This sounds like a small enough packageable project that has lots of moving bits, but like if Dato still can get us a place to post the results or if we end up posting it on this mass open cloud thing that Operate First team. The guy's name is Karsten Wade and Marcel Hill. They run that project within Red Hat and then there's Tom Kool is another guy and he's actually going to speak at KUKON about it, but I could pitch to them that they reach work with this the CRC group or just come to this docs group and maybe they would take on the building of CRC for us as a proof of concept that we can figure out how to do something and that way the community could manage on Operate First the build process for this because I was worried about broaching the subject of building OKD on this without having any experience and CRC seems like a small enough thing. Does that sound reasonable to people or am I? That sounds reasonable and Diane, you were not at the main meeting and you might not have talked to Christian at all, but at the working group meeting Christian said that there are internal discussions of something that might change with OKD that will make it more community accessible in terms of development. If that goes through then it would make more sense to focus on another project for what you're talking about. I've been listening in on those other conversations too and I hesitate to say anything because sometimes they don't come to fruition. We'll wait and see. We'll wait and see and I'm hopeful, ever hopeful, but I think that this might be a good project to introduce Karsten and Marcel and Tom into if people are willing and then maybe we can get a tiny iota of Garo's time to help them understand what he did and what the build process was is and then wrench the hosting of it, either leave it over on Fedor land or host it over on the operate first cloud and repoint things to where they should be. But this sounds like a good candidate. Because it's something that is really homeless. So there you go. That was my pitch for operate first. If you've never heard about it before, this is your first mention of it. Google it on Red Hat and you'll find them. They have a landing page somewhere. All right. Let's let's do it. I say anyone else on that? Pitch it. All right. We've got nine minutes. So I want to get through the the next few things that we have here. Twitter we talked about surveys. So I've messaged 3t and she has not responded to me. We've had that survey sitting around since November. I don't actually have the link to it anymore or anything. I don't know where it lives. But I'd like to get a copy and just have us push it out, even if 3t is not able to participate in them. So Diane, could you try reaching out to her as well? Yeah, I'll see if I can find her. She was everybody keeps getting subcombed into other projects. So I'll see where she's at right now. If we could just get a copy of it, then we could take over it. Because she was going to do the work but I think she's really busy. So it would be nice to get that survey out right around the next couple months because there's a lot happening and then we could get a sense of direction of where people would like us to go. And Vadim did give me some numbers. He's going to check on them but the amount of OKD usage sounds pretty significant, actually more than I would have thought. So this survey might give us a sense of who's actively using it and willing to put a voice to that. Then the other thing was bare metal testers. Do we put out a call for bare metal testers? Because that's another thing that keeps coming up is that people are asking about bare metal and have questions about bare metal and I feel like we aren't really able to answer them because we don't have enough bare metal testing. And granted that's a unique thing because bare metal is more unique than any of the other platforms. But there's some basic stuff that just we don't have documentation for or don't mention or can't support when the community discuss or help. Is that something, I know we don't put stuff out to the mailing list. Is that something we could, but it might be an ask is part of the survey or just an ask on the mailing list. Is anyone deploying to bare metal? If so, could you? Yeah, I think we can do it on the mailing list and then if maybe we could do something on Twitter that just says, hey, if you're doing okay, D on bare metal, can you shoot us a message or respond to this or something? Because there's folks that aren't on the working group mailing list that I'm sure are using. Okay. Is that the whole agenda? I had one other thing I wanted to sneak in. Go right ahead. Because Brian is in the UK. Brian, I am getting the agenda together for two UK based OpenShift Commons gatherings, one in Dublin and one in London. And I didn't know if you were aware of those, but I don't just as a FYI, don't call Dublin UK, they get really upset about it. Oh, okay. Good point. All right. So I don't know if you're interested in participating in either of those. They're in June and July. Certainly the London one, yes. And depending where things are with travel, getting the Dublin, but certainly the London one, yes. Okay. So I will reach out to you offline and you may be the okay D rep there. Just to do something that will be fun to make. Does that mean I can actually get a t-shirt now? You don't have a t-shirt yet, Brian? Oh, for Pete's sake. All right. Well, anyways, we will make sure that happens at some point. I'm time for a new run of t-shirts, I think. Okay. Cool. That's good. So that was, we have, I think July is, I can't even remember which, one is in July and one is in beginning of June, end of July, beginning of June, but they're far enough apart that you can't come just to one if you're traveling internationally and stay. That'd be great. Cool. And shameless plug for me in the chat is a link to, because I don't know that Christian Hernandez has a page that he puts these on, but I'm going to be on Get Ops Guide to the Galaxy on Thursday, talking about Argo CD on OKD and OpenShift. Someone who has it handy put the link to Charo's CRC build instructions so I can write that. Okay. They're on OKD.io. If you go to the CRC. Okay. I'll grab it real quick for you. Which one is it under, Brian? It's under the CRC subgroup. Oh, right, right, right, right, right, working groups, CRC build. I can see the installation stuff. I can't see the build stuff. I guess I'm in the wrong place. I think it's actually linked to that video. Oh, so there's a link to the video in the video. Is there the, I think I thought in the description of the video there was the, I might be wrong though. No, hold on one second. But if not, we should put it on that page. Yeah. Yeah. I can't find my own stuff. I forget what his, what his blog was called. Upstream without a panel. Yeah, yeah, exactly. There we go. Projects, stuff storage. He actually has one on deploying stuff storage. We could put two together and be like, do this, then do this. All right. Thank you. I just couldn't find it. And you know, I wasn't totally brain dead, but slightly brain dead today. Still, the flu. So we are just about, I'll put it on that page as well. So it's, yeah, that would be good. All right. Any last minute of things? We're about at time. Yeah. At some point we need a plan to work on how we get to the new get repo. Yeah. I was going to talk about that this meeting, but we had so much other stuff come up. Let's start formulating a plan. It might be best to just start just async, like just start like a document and just start a discussion. Someone's got to go to thought discussion thing that we're not piling on and let's get an agreement on how we're going to migrate it. Because what I don't want to do is have two sets of stuff going in parallel and everything, everything gets confused. So we need a certain cut over. Okay. All right. We'll start a discussion item on that. Brian, you want to go ahead and take the lead on that? Great discussion. Okay. No, I'll do it. That's fine. I'll do it. Oh, no, I'll just, I haven't actually got around to creating the other discussion I had to migrate to do last week. Well, I'll tell you what, I'll do this one. You do the other one. And then we'll split up the work. So it's fine. It's all good. We're going to get more people someday. I promise. All right, folks. Thanks for an awesome meeting. And I'm going to stop the recording here.