 And get started. I think I've got everything in here. Let's do a quick agenda review. Is there anything that you see that you want to change or add that isn't currently in the agenda? I think everyone has arrived after that was posted. Okay. Anything you want to add or change to the agenda? Let's move forward then with our first item. So. First item is the Charter update and placement. Nothing from me on that. I was working on some of the other texts actually, but that will be coming soon actually. Okay. And beta site Brian, go ahead and give us an update on the beta site. Okay. There's, I've not actually done any changes since last week. I'm not pushing any changes. I'm a bit busy this week, but I'm hoping to get the site to a point where at the next full meeting, we can have a vote on to put it live or not. So I've been sort of moodling some stuff around on my sort of local repo. So I've not pushed anything, but I'm hoping that by next Tuesday. There'll be the content there that we can vote to go live or not. And one issue has come up though. And where should subgroups be represented on the web? Because there's been a an issue raised on the OCD IO gate. And it looks like the virtualization subgroup have already created their own web presence. Yeah, I was a little bit. I, we had talked at the last main meeting about creating a sub working group. And they just sort of went and ran with it on their own and then created a presence. Do we want to ask them to, to reel that into what we have going on here as a most splintering. And that's what I was going to say, because I think we need to be consistent. So we've obviously got the new virtualization. We've got this document. So where do we, where do we want to put our, our content and the active talk topics that we're doing? And then I'm guessing that we may eventually create one about the operators are going forward. I mean, to me, having fewer places is easier to manage sending people off across the web. I think just gets over complex and then you just don't know how to update it. So I'm, I'm guessing they're using a different technology. They haven't actually said how they're, how that is created. There's no source there. It is a github.io, which is on a, is on a git wiki. I don't know what. I think it's a git wiki. Yeah, I think at the next meeting, we invite the vert people to come to the docs meeting and ask them to migrate into the mk docs. Let me, let me reach out to them though now so that they don't do a bunch of work. Do you know what I mean and actually say, hey, we'd like to invite you to meeting and show you what we have. We'd like you to move into. Yeah, I'm guessing the other side of it. I think there's quite a few of us agreed to join that group and we're not part of what's going on. Right. They did say something in the text of if you want to join us, come join us, but it wasn't very, yeah, I'll reach out to them probably later today. And I'll basically outlay, you know, hey, we want this to be participatory. Could you please use our web interface. We've just moved over to this. It's consistent as opposed to bouncing people around. And then I'll also give them the list of folks that had voiced interest at the last meeting. And I'll send something out to the group that will ping people again. And the other thing is that we want to make sure that they record their meetings and upload them to YouTube. So because that's our way of capturing and being open and transparent without having to do huge amounts of meeting notes and things. So, yeah, if you want to please do send it out. And if, because I bet there are a bunch of them are redheaders. So if you see, see me on that and they give you any gov. I think they just want to get it done and they want to jump to the chance. The easiest thing in the world is to create a web presence, you know, and that's like the first step everybody tries to do before anything else. And, and that's just, and we haven't gone really public with using MK docs and the beta site a lot. So that probably was confusing in anyways. So, yeah, the sooner we get to the MK docs and get out of beta Brian that the happier everyone will be as for where in the hierarchy of the MP docs and the pages this group lands. What I was going to do is I was going to put at the bottom of this sort of menu working groups and then each working group will get their own section. And what I'm also doing is thinking of putting a new section on the front page, which says current working groups with with sort of fast links to their landing page. So we're going to have three basically because we'll have virtualization, the operator subgroup, actually for CRC subgroup and also the documentation subgroup. So we've actually got four subgroups now that are coming together. So you realize this is actually the meeting time for the operators subgroup at the same time. Well, no, no, no, this, this, this actual meeting is during the operators community. Started out with the operators working group. And then we, and then this exact group with these exact people morphed to the documents group because not much was happening on the operators. Well, hey, there. This was like a year ago. I think that the real work of the operators and the operator framework and the OLM and all that is happening in a set in the CNC F now. I mean, it's, yeah, that's true too. It would be nice if it would come back to us. So the team actually gave me the contact info for the, the, the person who's heading up the project within red hat to release that subset into OKD. So that's going to be something. No, right. Would it be reasonable that these are all subgroups to the working group? Yes. Yeah, at least once in which it fits under our umbrella. Right. I mean, not, not the outside upstream ones, obviously, but the, you know, and then they could periodically report and things like that. That, that was my plan. My plan was probably not at the next meeting, the next meeting, but the main meeting after that is to start putting into sort of a consent agenda. The actual reports from the subgroups so that those get filed in with the main group. Good. Super organized. I love this guy. Okay. Anything else, Brian? Okay. Okay, great. Delimination of resources for working group and OKD users. That's my suggested text for Google. I checked with Google. We have to trim a few words off. We can do that. So I'll take some of the also's wills and things like that out and see if I can get it into the, it doesn't give me a number of character count. Keep trying. So I'm going to clean it up and I'll do that as we speak. And then we can look at it and see if it, it fits. Okay. And let me share my screen real quick because I want to show folks what I did for just do. Oh, no, actually what I want to do it. Okay. So if you go, folks should be seeing my slack. So you go here. I did, I don't know if anyone noticed this, but I did change the text on the channels over the weekend. If anyone has an issue with this, let me know. But this is what I came up with discussion of OpenShift OKD development. Please refer to OpenShift dash users for installation and usage questions slash issues for OKD specific help visit the community led discussion forum. So there's that. And then on users, I did this discussion of OpenShift installation usage questions slash issues for OKD specific help visit the community led discussion forum. Does that work for folks? Anything? That's what I'm, I'm trying to trim down. It's a little long. So for the, for the depth one. So this is, I mean, is it okay on the slack channels at least on the slack channels. It's fine. Okay. That's, that's fine. Yeah. Okay. So slack channels were good. That's, that was something I wanted to run past because I didn't really check with anyone. I just sort of did it. Okay. And then, yeah, then for the Google group, Diane, you're going to trim that down. Name and scope of the install.md and readme.md discussed at the main meeting all were in favor. And again, for folks that were at the main meeting, Vadim just said, you know, yeah, I think this is a great idea. Just make sure you link to the proper sections in the main site so that people can see that. People can get to all of this information. That's easy enough. So basically, I think we're good Brian to start like moving things over to that, like, you know, just start separating things out from install md and readme.md. And we can point, if you want maybe a future docs meeting, the next docs meeting, if we can just walk through people through where it got put so that everyone can sort of know and point folks to if it isn't obvious. Does that seem reasonable? Yeah, so I mean, what I was going to do is I was going to do it, create a new readme for the OKD repo. I'll not do a pull request. And so I'll probably show it at the next meeting. I'll put a link to my Git repo where it'll be so people won't actually go and review it. They can go review it. I'll probably put something in the Google group with all this in the meeting. People got links to following things. But the idea is at the next main meeting, the mkdoc site will be what I think we can go live with. I will have updated the readme in the OKD dot of the OKD main repo referencing the beta site. Organization. And then we can also close the next one just to save time because that ticket will no longer be valid and we can actually make sure that the new site is fully inclusive and passes all the scans, etc. Apart from as we go live, we've got this complex set of actions to do in terms of disconnecting the Red Hat hooks to the current build system, renaming the master to main and then moving, obviously promoting the beta to the main branch and then getting the CNAME registered and whoever our provider is getting them to update the pointers to the OKB. Let's write these out. Let's go back to beta site. Let's do tasks. Let's write these out. This way we know that we get it done. What are the tasks? It's unhooking the web hooks that link into the current Red Hat build process for OKD.io. There's a couple of web hooks in there and I'm guessing there's some back end stuff to clean up as well. We've then got a rename master to main for the inclusion, inclusive language. We've then got a promote beta to main and then we've got to whoever is the DNS provider for the OKD.io site. We've got to point to the obviously the GitHub site and then add the CNAME to the GitHub pages. Anything else? I think that's it. Excellent. Oh, thank you, Brian. Something broke here. That's me. I'm sorry about that. I'm just. That's right. I can read through it. OK, there we go. Perfect. Inclusive language. OK. So I don't know if other folks have had a chance to look at Michael's document. I was looking at it the other day and to me it has everything we need. I can't think of any changes. Michael labeled it first draft, but. This is perfect. I can't see anything that would change. Has anyone seen anything in this that they would want to change? If not, I say we ship it and just put a page up with it and the beta wiki. Yeah, I'll do that. Yeah. Great. Michael, is there anything you wanted to add or change about this? No, I don't think so. Thank you for your help on this. This is really awesome. We literally get questions about this. We have a ticket that came in that was that was needed this information. Actually, and it's it'll be helpful to have it on a page that we can. And as things change, feel free to let us know. Right. Right. What do we have? So create a build doc outline. We're going to hold off on that still. Vadim is out for 2 to 3 weeks. He will be back theoretically in October at which time the site will have moved over. And so we'll be able to go with this build doc outline. And so we're sort of on hold now, still with that. Upgrade path notifications. Bruce, do you want to talk a little bit about the conversation you had at the main meeting with Vadim about this and what your takeaway was? Okay. Well, I guess my. Wait, wait, wait, wait. Okay, good. I'm going to stop sharing and then Bruce, you can, if you want to share point anything or anything. Well, no, I'm just going off the meeting minutes there. So, so basically I had a question for Vadim on what the, if the upgrade path did not have success. But it had S's and F's what that was. And my guess was totally wrong. It turned out that those are just specific tests that had succeeded or failed. Which I'd already discovered that you could find by clicking on them and go through the pages upon pages of what they were. So I had just been imagining something more than actually was. I don't know. It might not hurt to document that. But I think that that actually uncovered a. Bigger issue. Which is that. I'm not sure that we really know who our audience is. You know, I sort of generalize it tremendously there. And more specifically. Audiences. One is people that want to use this in production. One is people that are using it. To see whether or not they want to use OCP in production. One is sort of hobby people and. They have different needs, basically. So if you're wanting to use it in production, then you are concerned about stable streams. That are actually stable. And we've seen that even the F cost stable stream can wreak havoc on occasion. So that's not really in my definition of stable stable. And the. I actually upgraded my test cluster to the latest and greatest and was pleasantly. I wouldn't say surprised, but what was, you know, I guess pleasantly observed that it worked. So I've gone to where my underlying assumption is that it's not going to work. And so if it's something that I don't want to take the time to try and debug, then I will skip the updates. So, and then related to that, the somebody asked, okay, well, they're requesting the documentation and then Vadim said, oh, well, our documentation is for 4.8 because that's the latest, which is something that we haven't released. So it seemed odd that our documentation tracks something that we haven't released. Yeah. And I so I actually have this actually dovetails into the new business section. Right. And I've got a link to that GitHub issue. This is issue 882. And basically it started out with redeem Vadim said, oh yeah, so it's in the latest release. And someone said, well, I don't see a 4.8 release. And I said, well, he means nightlies and they and then they said, well, okay, then you should make that clear. I agree. This this all sort of dovetails into the same thing as we need to be more clear about what's a nightly release. What can it be used for what's a real release and that one release page with the tabs and whatnot that's available on that CI server is nice. But if you're a user of OKD or hoping to be a user, baby, OKD navigating through all of that, I think it's probably not the best, right? Right. Well, you notice that our documentation goes sort of like one point something and then up through two point something. And then finally we jump to Kubernetes and it goes three point something up to like 3.11 and then it's latest. And we also get people that are trying to install 4.4, 4.5, 4.6. Yeah, I saw that actually. So it might be useful to have actually like we have gone through documentation for all those things. We sort of never separated the documentation by version. Right. And again, I think that's the underlying assumption of production versus latest and greatest versus development. And maybe it all tracks back to resources. OK, but I think we're just sort of all into one big mush without clearly delineating the different stakeholders. And sometimes that bites us. If I might ask a question of Michael Burke, I think a few meetings back, we asked if you could generate the interim four point releases so that we didn't have the latest. What was the status on that getting that done? As far as I understand, we can do that if that's what we want to do. Just need to know exactly what we want to represent up there. So, let's take this back to the main group meeting, but my sense is we're probably going to want to go 4.6 and up. Yeah, right. And and latest can't be 4.8 because it isn't actually our latest. So I'm hoping that there's a way to tweak that pull down to make it clear that latest is 4.7. And maybe we follow sort of an F cost model that 4.8, which we can't release yet is considered next stream like an F cost or something like that. Or we give it a name that indicates it's on its way, right? Or you just even call it instead of latest call it next. So you have the latest, which would be 4.7 and next would be 4.8, but I don't even know why we would list the next. Well, anyways. No, I think that's a valid point, right? So, or we would call it testing or something like that testing version. Well, are we actually tied in with OCP here? Or can we actually have different versions in OKD or are we actually in a published cycle with OpenShift? Because I'm guessing by the time of a Dean comes back 4.9 could be out. That's the thing is so because there's one channel basically for all of the minor releases, it's always been that we sort of have our own release schedule. And this is confusing because actually someone asked a question the other day about that new security bulletin that came out the other day. And they wanted to know what OCP is addressing this is F cost or is OKD addressing this. And we actually don't have any information about our addressing of security concerns on anywhere on our website. So anyone that wanted to use OKD for production, they wouldn't know like, okay, this is we are incorporating updates that fix this and whatnot. And this incorporates such and such bug fixes and whatever. We aren't beholden to them because we are built off of F cost. And because there's that underlying issue that's preventing us from doing or eight. We won't be doing for eight at least for like another month or something like that at least so I mean like an official release. I mean, the question comes are we going to skip for it and go straight for nine or but and I also think from a document partition point of view, not just OKD releases, but also the documentation releases. Are we tied in with OpenShift OCP. On the dock site or could we actually have different versions under OKD than OCP has. Yeah, Michael, would it be, would it be possible to have a like you called them books before, would it be possible to have a different sort of top level. That's the same style but not based off of an OCP like 4.6, 4.7, 4.8, etc. Or is that like too much work and to do do you think. Well, but that's our numbering right like we do use that same numbering and even if you even if you're getting a beta version. It's some number, you know, 4849. You start talking about 410. So it seems like it would be handy to have documentation that is tracking whatever OCP is doing, which is the basis for hours. So if you were using, if you were trying to do like a 4.9 nightly version, it would be nice to have the 4.9 documentation. Now, I mean, you could just say, oh, I'll just go to OCPs and fair enough. Yeah, if we want to be a product in our own right, I don't think that's sending people the OCP documentation. You know, or, you know, so Michael, I guess then the question would be, so the 4.9 documentation is it's OCP 4.9 documentation. Is it currently available in a like a beta form? It is available, but not publicly available. Okay, it's not publicly available yet. Yeah. So, Michael, in in the OCP documentation, where did they for security updates? Where does that that do you pull that in or did you just link back out to the security section of the red hat page or patches would be in the release notes. Yeah, they don't actually, they include it only in release notes. It doesn't get put into like the, the, the booklet says Michael was calling and that are part of like the installation and maintenance docs. That doesn't contain any like timely update information generally. And does that surface anywhere in the release notes that security stuff doesn't know we've never taken stuff. No. Yeah. No. Yeah. So, I'm not sure what the group is landing on here. It sounds like we ought to have some, I hate to keep creating read me dot md pages and things like that, but like some notation about how this scenario plays out. I'm going to go short right up on that. And where to go to get more information, because we are reliant on the OCP patches. There is a what's new section in the documentation. Yeah. So, what's new is not quite the same thing as security. Symmatically, I think what is so what is under the what's new section now in the beta site. Let me go to that link again. I always have to look for the link because I for some reason. I don't have it memorized yet. So it's in the chat. Okay. Can you put it in the meeting docs so that it always gets copied when I copy over the meeting for some reason. The new site. Okay. So, okay. So this is the old right. So this is the new features. This is basically version to version stuff. Right. So this is basically like what's different between in here. Let's share. Let me share my screens. This This is basically stuff like what's new from version four, five to four, six and stuff like that. I think. Or is this like four versus three. Four versus three. Yeah. I know I haven't updated it. Yeah. Michael, do you have the history of how it just became like okay D for like a generic. Maybe you don't know the history, but sort of how we ended up moving away from like okay D individual sort of. You know, 4.6 4.7, et cetera, et cetera. So that it just like one of the things that someone asked, I guess I should rephrase this. One of the things that someone asked is that. You know, there is no 4.6 4.7 individual documentation. And if we go here to documentation, it's not in the pull down and you said you can add it and generate it. Do you know the history of why it. Stopped at version 4. And just jumped to. I was told. Quote, because that is what was agreed upon when we came out with version 4 of OCP. Okay. Yeah, that's probably. I didn't pursue it. Yeah, I don't think there's much. Yeah. Michael, I think it's probably more on me not insisting on it and not realizing the implications. Yeah. I mean, I don't know. I don't know. I don't know. That way back when and for that and just being grateful to get any documentation generated at that moment in time. Yeah. I'm sure it wasn't laziness or not wanting to. I think it was just. Me not knowing exactly what to request. And then. You know, until now, nobody's really asking about it. It was about a year before between the time that OCP came out. And we actually got a release version of okay D. Yeah. And what was our was it 4.5 or 4.4. I forget what the 1st 1 was, but what we should probably do is go back to whatever the 1st 1 was and figure out what that was and have all of them was. I think that would be useful. Yeah, I think we definitely want to do that, especially if it's okay. Okay. We could revisit it, but. Okay, because one thing that's always kind of annoyed me is if I do a. A web search for a particular topic. I'll get the link for like 4.3 of OCP. You know, or 4.1 and then you get that big red text that says, you know, this is not the most current documentation. For some reason that annoys me, I think only because it ends up being rising to the top in terms of web searches. So that's that's my only concern is that if the old documentation is always rising to the surface that could be problematic. But yeah, we say we go ahead Michael. I was going to say you get that with the documentation. Whenever I'd Google I was ended up with a, an old version of it now to go and click the latest version and says this page isn't available. This is what OCP have. This is their drop down. I'm not sure why this hasn't occurred to me before but. The organizational plan within red hat. Is it docs.openshift.com. Website is going away. We all done to our customer portal. I'm not sure what that means. Okay D docs. Find out. The openshift docs aren't going to change their URL. Is that what you're in? So docs.openshift. Plan is docs.openshift.com is going to go away. Go away or at least redirect. Yeah, I think it's a redirect so. But it's still going to be generated and going to be publicly available. It's not going to be behind a firewall. So, as long as ours keeps going to doc from the same batch. The output goes to docs.okd.io. I don't think that. Does anything to us. Is the technology behind the docs generation change or this is just a it is okay. Okay. Really. Again. And so what is this? Can you find out what the impact will be on this auto generation between. Whatever the new system will be and we're going to want to find that. Yes, I can do that. Okay. Thank you. Add that to the agenda for the next docs meeting. That's really that's that'll be fun. That'll be fun. So, are we saying that we do want to go all the way back for okay, these 3.11 3.10 3.9. And, and have all of those regenerated. Or do we want to just focus on what's realistically going to be supported anytime. Well, I would just say the ones that we released in the four slot. Have you got any users? I seem to recall on the forum that there's been a couple of posts where people said they're still running 3. Yeah, there are still users out there of the 3 dots. In which case we need to keep the documents available for them. Yeah, and you can't change them on them at this point in time. So you can't like regen with the shortened version that would just drive people. The terrorist mode. Or me. So I think we need to keep all the 3.x up to the latest, but drop all the ones and the twos. Which is what, which is what red hat open shift is done. That seems reasonable. Yeah, I go with that theory. Michael, what are your thoughts on that as the documentation expert. That seemed legit have three, three X and four X, but drop the two in the one. You're muted. You're muted, Michael. Now, you know, the user is better than I do. Yeah, there are no 1 dots left anywhere in the universe. Yeah, so I would say, yeah, if you could start with three, then that would be great. So three, six and then up. Do that. I guess. And do we have a time because that will affect the beta links. If it's going to happen, if it's going to happen quickly, then I'll get it good for launch. If not, when it changes, I'll change the docs. Okay, I will look into that too. Thank you. I'm just going to pause. I'm going to turn my camera off because I'm in another meeting at the same time and I have to talk now. So, okay. All right. I'm keeping recording. Keep going. Okay, so there's that. We talked about that. And noting release version in the docs. We'll also want to figure out a page that better describes nightly. So this is something that we're going to have to write ourselves or have a team right or something like that. And incorporate it into the website. And so that should explain the nightlies process. And that, you know, nightlies are, you know, pruned after 72 hours and basically better explain nightlies so that people understand what they are really, you know, and so if someone wants to take on that task. That would be helpful. I think we need to recruit a couple more docs group people. Maybe we'll send something out and say something at the main meeting and I'll send something out over the email list if we could have one more person to volunteer time for docs. So particularly docs that we need to written new docs that would be helpful. Yeah, I mean, we did tend to have an extra couple of teams have dropped off in the last three or four meetings. Yeah, I'll reach out to them and see if they want to rejoin us. At one point, Diane was promising promising us hordes of newbies. Right. That's right because there were supposed to be there was this horde of new red hat employees that were like they were supposed to like, you know, basically start out here to get their wings tempered. Right. Neither they ran off or they did so well that they've moved on to the OCP stuff. I don't know. Okay, so let's put that as a task item. I'll take care of that. I'll reach out to the people that used to be here. And then we'll also send I'll send something out over the mailing list, the Google group to try to round up more people for the docs stuff. We can have one more person who actually writes new docs. That would be fantastic. So this one came up that's interesting is that change logs are missing. And so if you were to go to I need to share my screen. So, Change logs are missing. So if we were to go here. And, you know, these are the index of releases if you go to 47 stable even the one from the other day, you get this error and says could not load and this is Vadim's repo. Right. And the funny thing is if you go to Vadim's repo. There's like no commits here. It's beyond like February 17. So I'm, I'm confused as to what's happening and Vadim is off for a couple of weeks and drop down the branch. I think bills are now done in branches. Is he doing. Oh, okay. 47. Yeah, it's kind of hard to tell which one it would be is it for 47 are back the one develop. Search right by. Mid. Use a new feature just hit a hit period. Where, where, where, where here just on that page just hit period. Oh, I know you can search. Oh, up top here. The left hand thing. Oh, right. Oh, right. Okay. Well, so yeah, I mean, I was not able to find it when I searched for it. So this is an issue for all builds. We won't be able to do anything about it until Vadim gets back. But if you hear anyone mention this, it does stop at, I think it's 325. Yeah, is the first one where it fails. February 14th works everything after February 14th fails. I'm assuming it's not branch related because these are all the same branches basically it's all off the same branch I think that gets sent to see I so we'll have to find that out sort of what the deal is with that. The other one that came up is related to. I visited this in years. There is a documentation error where in the OKD docs. There's a link that's supposed to grab some YAML files for Ansible some basically some Ansible playbooks to install. And that's a virtualization. And those failed. There is a ticket that was opened for OpenShift docs Michael at 36617 that someone opened up. And so we will want to fix this apparently the issue is that the it's particular release versions are accessible, but not just like released for version for anything. Like here's release for and apparently that will fail now it might have in the past, and it actually needs to reference the minor version. So Michael you have some work cut out for you there on that one. And this is I'm not sure how. Oh, he did put OKD only in the in the subject for that. I mean, I've also raised an interesting question should we be linking to Rev or Overt the upstream free community version. Not if we're talking about so I think that this is problematic for a couple of reasons. Let's have this discussion at the main meeting. Let's put let's earmark this for the main meeting, because ultimately this is also a technical discussion, particularly as we start talking about Covert and stuff like that. And I and actually I notice actually that the Sandro chimed in on this conversation. So let's bookmark this for the main meeting I'll put that as an agenda item. That's a great question, Brian. And something that we definitely should resolve, particularly if we're talking about this, this shift. And I'm looking at the documentation and the menu does say overt. So where where is the link that says Rev. OKD installing. So it does actually say overt at the top not Rev. Right. So yes, we will earmark that for the next main meeting. I think that'll be a great conversation. We can get, you know, Sandro and other folks involved in that as well. And I'll mention that. The other one that we had here, there was one more. Oh yeah, so Diane did get us another office hours during coupon North America is going to another live stream. So 1013 to 230 PST, which is, you know, 5 to 5 30 Eastern. And I don't think we've picked a topic yet or or if we'll just riff on questions that we get. Does anyone in this group have any suggestions for like a an area of focus or anything like that. There was mentioned that the arm version may be available by that time. But there's no guarantees. So that might be something that we better riff on that. But on version of what. So the arm version of open shift and by proxy OKD. Because I'm guessing Vadim is going to have to create a new installer. He will now how fast he does that. I don't know. And this is so. This is going to be brought up at the main meal. Bring it up here because it should come up at the main meeting. I'm going to propose at the next main meeting. That we get multiple working group people. CI access to do build so that. Vadim is not the only point person for OKD. I realize it's his baby and maybe the larger group will disagree with men is but I'm going to propose that we at least have one other person or ideally two other people. That are point people for building OKD. So like for example, he's gone but he said he would still do the builds the next two weeks or three weeks since it's automated and not make the desert. That's cool. But like what if Vadim wins the lottery or, you know, you know, and it's just I think it would help to have multiple people familiar with the build process and Brian this goes back to sort of the discussion of documentation on. Do you know are there any red hot resources involved in this, or is it purely within Vadim's home at the minute. If there is some red hat resources, which is the CI is a red hat CI that uses sentos. This is probably going to be red hats as then. Yeah, I maybe I don't know if there's a way. And if it is problematic and maybe we need to come up with a separate CI. Right, well I would imagine that you'd need access to the red hat intranet to get access to the machines. It's probably has got to be red hats. Yeah, well, yeah, and Diane might have more info on this but it seems like if anything if it has to be more red hat people, more at least more people because right now it's literally just Vadim. And if we really want to take OKD to the world as a solution for production that can't rest on a single individual. I just don't think that that's. Yeah, I agree with you Jamie and one of the things that keeps coming up is being able to build OKD from scratch outside of the red hat environment. And the sticking point often seems to be the machine config operator. According to Vadim. So that's that's been long sending I guess ask. And that sort of relates to what you're you're doing even if like if people could build it themselves. Then you'd have sort of more survivability. Even if they couldn't build official versions. Right. And I did toy with the idea of setting up a CI for I probably will in the next week or so because we're getting a lot of questions on single node OKD. And it seems to me that single node, a CI that tests single node would be a really great idea because we kind of got caught with our pants down. Vadim was convinced that single node worked. And it actually doesn't like 46 or 47 and it looks like a version of 48 it's broken as well. I actually think we need to actually review. The whole point of automation and testing because obviously we've got the CRC group as well. And we need to automate that. And then several times Vadim is sort of alluded to a lot of the instability is because we don't have sufficient test capability within the current process. So it is something that if we as we move forward I think we're going to have to look at across the board for all versions and all platforms of OKD plus the CRC. Yeah, I think the CRC pretty much covers you on the single node because 90% of the work of doing CRC is building the single node. And then you scrape it down, you delete some operators and then you package it. But it's a very different process to the OpenShift installer isn't it because that was the 34.8 SNO version wasn't it. Yeah, right. Yeah, I mean for 4.8 was supposed to solve the problem. Because it was a feature in 4.8 single node. But the installer version does it didn't work for someone. The other day and but I was able to get it to work so I don't I don't know that might have been a configuration issue on there and then this was in AWS IPI. Brian and then Bruce. We need to put this on the agenda for the next full meeting. And because obviously there's documentation to be done. Then there's the whole CI infrastructure question and getting more people support but also extending it with the new use cases. And then potentially we need to get Vadim's Git repo into somewhere a bit more official. Yeah, agreed. So I will add that to the agenda for the next meeting. I think that's very important in terms of survivability of the project and also. You know, being able to publicize the project and being able to have confidence that when we say to people that something works. And a relative level of relatively high level of confidence that something does work like single node installs or, you know, or whatever. Right. So, okay, that will go on the next agenda for sure. We mentioned the OKD office hour. Here's the sig stuff. So as we talked about, I will email them and ask them to move their stuff into the web. So, Brian, I'll CC you and I'll also CC Diane on that. Just to say, hey, you know, we have this new web system. We prefer that it stay within this and also here are the working group members. And hopefully that that will just go smoothly and then Brian, you can show them next. Yeah, I can say if you can send that note tomorrow, I'll make sure that tonight I actually get the landing place in the beta site. So you could then put a link in. Perfect. That would be awesome. And what was the other thing I thought I had one more thing that I wanted to mention. Oh, I know I remember. Twitter. We need a Twitter account. This came up like right when I joined the group. You know, year and a half ago or whatever it was or a year ago. We need a Twitter because so much can be so much promotion is done on Twitter. And when we get a new release, no one really knows about it because we're not like, oh, OKD, you know, 4.7. But, you know, September 21st version came out. Check it out. New features bug fixes. We don't do anything like that. We don't promote ourselves. And I feel like that's something that would be really beneficial to people to us increasing adoption and also getting people using it so that it gets tested and it gets improved as well. So anyone have any thoughts on that? Well, currently we point the bottom of our documentation points to the open shift Twitter account. Yeah, which never includes any OKD stuff never has never. I predict never will right because they have that sort of interest in every open shift so. Just to ask Diane, but I'm sure there's some sort of branding type thing. You know, but if we could just do like OKD Kubernetes or something like that, like, you know, and just get a one of the newbies or something or a couple newbies or some of us to staff it right into send stuff out. I think that would be really, really awesome. I think that's a good idea. If you could, if you could also get, which I know Diane I think has been working on if you could also get some of the. All the the people are doing open shift demos. If maybe a 10th of them would be on OKD that would probably be helpful too. Right. Yeah. I think have Edison doing OKD demo. You know, in like your script wouldn't have to change. Right. Okay, I think that's the agenda we have like 30 seconds left so let's go ahead and call it. See you all. Everyone here except for Michael is at the main meeting. I just want to ask one other thing if we're doing Twitter, do we want to do Facebook? Because again, we point to the Red Hat OpenShift Facebook group. I think yes. Anyone have a disagreement with that? Anyone think it would be a bad idea to have an OKD Facebook? I guess the only issue is if we have these things, you've got to do something with them. Right. That's absolutely and that's where the newbies come in and us seasoned professionals who have some social media experience. And we can also automate some stuff, right? So if there's a post that goes to, if there's a commit, you can do things where like a commit made to repo then calls the Twitter API to make a Twitter post that, you know, has that information. So we could look at doing that as well. At least from the perspective to a SPOM feed because of that. Absolutely. Yeah. Yeah. All right. Well, let's go ahead and stop it there. And thank you, Diane, for offering us this great conversational opportunity to make OKD better. All right. Perfect. I will stop the recording. Jamie, if you have a few more minutes. Just want to catch up and then we'll go on from there.