 and share my screen. Welcome everyone. This is Doc's office hours. It's the 10th of August. And let's look at what we've got as proposed items on the agenda. Let's see. So I think I may want to add one item, Jonathan. If we get through the topics that you had, we should talk about, oops. Oh, that's silly. Why won't this thing format? There we go. Let's see. Windows upgrade guide for Jenkins 2.235.4. And we can just discuss it briefly. Okay. So here are the topics that I see. Welcome, Vlad. Great to have you here as well. We've got Jonathan has done some really excellent work on looking at the wiki migration sheet and has turned it into something much, much more useful, much more interesting. So that's the lead topic. And then I put Windows upgrade guide for Jenkins 2.235.4 as a second topic. Any other topics anyone wants to add to the agenda? This looks fun. Okay. Then let's start with the wiki migration sheet. Jonathan, do you want to take us through a review of the data and I'll navigate or what would you, how would you like to approach it? Well, let's talk about the topics first. There is some, all data, it's from the wiki migration sheet. So it's no secrets for anyone. But I believe now you can see a graphic visualization. We have understand more about our work we need to do. So for example, I discover we have several pages already migration, migration, but some of the redirect links isn't work, maybe cause the redirect instructions were overriding for another comment, for example. I don't know why. So we need to investigate why the redirect doesn't work. Do you have an idea about the issue? I do and it's at least one of the issues is due to a mistake I made in creating these links. So let's take one that has an example. Yeah, first sample. Let's choose, let's see, I should be able to find one that, oh, okay. Installing Jenkins on Ubuntu, this one. Yeah. Right. So when I click this, we're going to, it's going to take us into the wiki, but it's going to add an extra slash character into the URL. And that was my mistake when I initially created the link. Notice this extra slash character up here at the top. So instead of being Jenkins slash, it's Jenkins slash slash, which is still a valid URL, but it bypasses the redirect. So what we need to do is we need in this, in this sheet, we need to, and I don't know, is it possible to do a global search and replace where we turn every occurrence of Jenkins slash slash into Jenkins slash? Okay. But if you want to do that, are you positive that nobody had any of them set not to redirect? Well, well, so go ahead. So I can, Meg, I can answer your question. Yes, I'm confident that this was an unintentional thing. And any case of Jenkins double slash should be Jenkins single slash. So I'm confident the replacement is valid. But Jonathan, to your question, what was your question, Jonathan? Well, it's because so maybe the issue was writer with double slash too. It also was. Yes. Also, and that, and so the GitHub issue was also wrong. I made the mistake. I was consistent in making the mistake. Yeah, you're really good, though. Yeah, yeah. It's what happens when I use that. When I use programs to create bug reports, it's disastrous, right? And that's why I didn't create copy pastes of them. But yes, I was using a program and the program helped me do the same thing over and over again in exactly the wrong way. But I would mark that consistency is very positive thing, at least positive here. That's the way that's the way you have great potential. I agree. That's good. It's really dark, but at least it's dark everywhere. That's great. Okay, so about the redirect signs, we have no issue about just a mistyping. No, actually, some of them still are not redirecting correctly, because even with this change, I found at least one case where the redirect was still not being applied. So your done needs redirect comment is still valid in many, many cases. And we have to visit it and see, okay, is this one of those cases or not? Okay, and about the voice lashes, you want to put another rule in the Vrocious or just omitted from the weak migration sheet? We definitely do not want another rule in the V host configuration file, because real human beings don't make that double slash mistake. So I don't want to, if you've looked at that V host count file, it's enormous. And I don't want to make it double length, because I made a typing mistake in creating bug reports. Nice. Okay. So this week, I will check link by link, looking for double slashes and it's really not redirects. It has no redirections. So we can map in all the works. That's correct. The double slashes do not have a redirect. And in fact, there is now a safety check inside the Jenkins file for the vhost.conf repository, which checks that double slashes do not exist for this case, because we had, it's my mistake went even further. A few of the redirects were defined initially with double slashes. So they were failing to redirect because they did exactly what I said to do, instead of doing what I meant to do. So any objections if I attempt to do this replace? No, it's okay. Okay. So values on all sheets. Yes. Okay. It's okay because I will use the status column to looking for the not redirects. Great. Okay. So done. Now let's test it just to be sure that it actually did something. This should go to the wiki and then, so apparently the global replace thing that I tried didn't actually do a global replace. So we'll have to find another way to do it inside the URLs because what we're trying to do is do an edit of a URL, not edit of text. Maybe it's because we use a hyperlink formula to create a hyperlink, but there is no problem. I can fix this using another formula. Okay. Super. I know I can do the edits interactively, and they then work, but yeah, I don't want to do the edits interactively if I can possibly avoid it. So that addressed the embarrassing one. Now the recent change, so we just had, just within the last week or two, we had reverse proxy brought in this one. Jenkins X behind an engine X reverse proxy. And I know that the change was submitted because I submitted it. I was one of the reviewers, and it's been merged. And yet when I take away that extra slash, this will not redirect. So, but it should. So there's something still amiss. And this one needs needs more investigation. Okay. So I'm just going to put a note here. Jenkins slash slash Jenkins slash is many of the issues. And then the proxy reverse proxy recent redirect is does not seem to be having effect yet. I know that it was merged. And I'm going to put the URL there just to have it. Okay. Well, the next one, it's about the abandoned issues. So in that moment, we have a moment, we have a 12 issues waiting for PR reviews. Okay. But some of them already was reviewed by someone that asking for some adjustment, some mispiping or things like that. But the sender just abandoned the issue or didn't see they required some action. So maybe one of you with permission can interview and the approval of the PR. So, and this is one where did you did are these were cases where you detected that they hadn't made the changes, but or they were there are still changes that need to be made or both. But because the review asking for changes, but the sender didn't see abandoned or any case. So we are in the backlog. We can move forward because we are we need to interview from the inside of the PR approval, the changes or make the changes they review itself itself. So we can help us with these issues. I believe so. I think any any person any person can propose changes to a PR. But I think only reviewers. Reviewers can apply those changes. Yeah, we did take some actions that we need to choose what to take. Now, is that did you did you find a query in GitHub that helps us identify these or do you just worked through them there only about 30 total open PR so. Well, I visit the issue by issue. So if you go to Wixit, we can use the filter to identify them. For example, all right, well, let's look then let's go there. Let's look at Jenkins.py. The second tab. Your second tab. Oh, you say here we've got them. Oh, okay. Yeah. Yeah. I put a filter there so you can clear and just clear and a PR review just PR review. Oh, good. Okay. Got it. All right. Okay. All right. So I can use this to drive do a quick quick sweep through things that we could open them up. Excellent. All right. Great. Exactly. And so, for example, let's see the issue. This 17 line, 70 line, here. Okay. Yeah, you're so I put in the comment that the PR rather prove it was abundant. Yeah. And this one this one is is a the glossary is a place that we have to be very careful about allowing changes. So yeah, this one we can bring it back. I'm confident we can we can re raise this one back for discussion. Let me do a quick look to see what the content so it's the PR 3362. There is a comment from Megaton. Okay. So 3362. Good. All right. Yeah. Okay. Just to visit the PR, we see all I get a request changes but wasn't done. All right. And it now has conflicts. And this is one of those where yeah, so this is this is highlighting the changes to terminology can be quite sweeping. And right now the governance board will meet this Wednesday or this Thursday. To review discussions about terminology. So this file is going to be touched in in context of terminology. So this one is likely going to hold while we while we work through the terminology discussions. Yeah. Okay. It's because I have no comment. I like about the talk. So maybe it's important to register by common to everyone knows about well and and it should it should have had. Yeah, there's already a PR and here's the link to the preceding PR. And yeah, so terminology, knowledge and discussions are in progress with the governance board. And we'll resolve in other changes in this file. And then the conflict. Yeah. So all right. We can actually, I mean, if we want, we could go through each of these and identify which could be quickly, which which could be unblocked if you'd like, or we can go back to the agenda. What's your preference, Jonathan? Well, Matt, I prefer back to agenda because we have no so much time to the meeting. So maybe it's a homework to do after. Okay, good. Mark home of quote. Yes. Yes. Okay. Next. No. Okay. And then next topic is about PR that were approved and measured by issue was not automatically closed. So we need to help to close it. There is a filter to them too. And so you can see on the sheet, just apply the filter there with sheet done, but not close it. Yeah, over there. So we've got just one there. Okay. So this one has been closed? Yeah, the measures. Yeah, you measure the PR, the close of the PR, but the issue is open yet. Okay. All right. So then, and do you know if this one has been redirected yet? I really don't remember. Okay, let's check. No, it wasn't. Okay. It needs a redirect. But there is double slash. Yeah. No, it needs to redirect too. Yeah. So let's fix this redirect, redirect instructions. And I'll need to find, so this is another homework item for me. Good. Okay. All right. Okay. Okay. Next topic. And then I'll talk about the issues. Okay. In the yellow, if you see the graph, the chart, the yellow ones is pages of aging issues. So I open someone, I open them. I think didn't, didn't we need to talk about this one first, awaiting PR review? Oh, I've already been talking about it. It's just for someone, we have 12 waiting for PR review. So I don't know how, yeah, are the times that the reviews. All right. So this is just a matter of someone needs to review them. Okay. Got it. Yeah, just above. Okay. It's good to know how page are waiting for reviews. It's just like a control. And the yellow ones, it's about the pages waiting for issues. So can I open all issues? I already opened some ones. There's some problems open a lot of issues at once a time or not? Well, the problem, we prefer not to just open issues. Because when we open an issue, the issue, it's not enough to actually open the issue. What we need to do in stating in the issue is here is the page that needs to be converted. Here is the destination where the conversion should go. And here is the content which needs to be discarded during the conversion process or completely reworked. So automated creation of the issue is not a significant help compared to a full triage of the issue, a full review of the issue where then someone can pick up the issue and know where to put it and which things to discard. So the new issues you created, they follow the same pattern that my issues had, which was not telling people where to put things. And the problem with not telling them where to put them is they cannot retrieve the issue and work on it because they don't know where the content goes. Yeah, but about this strategy, we aren't going to tell the people that we need help with that integration. So I believe if we have the issue, I can alert that I work on it. So we start doing the triage. For that reason, I link your Hackfest video tutorial about the immigration. And there, inside this video, we explain how we need to work about the planning in relation, how we need help, how asking for the nation of the new page. Right. So what you're saying is like this, however, what we're missing here is where should the post initialization script be placed? This talks about, okay, how do we do a migration? What it doesn't tell them is where in the destination should this thing be placed and what content from it is useful and what is not? Okay, so if I visit one by one and create a plan for migration and update the issue, so we can register of them? So can we register? So your question was if you visit each of the newly created issues and add to it, this belongs here, would that be enough? That would make them much more useful. So are you comfortable that you are confident you know where to put those kind of things? So this one, for instance, launching agent jar from the console. Well, I just add a proposal to destination. So I will show to you, for example, I guess this page about this list of jar from console needs to be redirected to that link, that new endpoint. So if you approve, we register the issue. Yeah, so this specific example is a really good one to highlight. Okay, we've got a page that has no useful content, right? It's just a link to another page. So this page is just a redirect, right? And it redirects, it's a wiki level redirect and wiki level redirects to this page. And this page, I suspect we can find this content already existing on Jenkins.io and just the action for these two pages should be just do a redirect to the one page that talks about how you launch an agent from a console window. Okay. So the analytical work, the let's see agents, the analysis to decide how to do, how to find those is very labor intensive, right? It takes you or me or somebody reading through those and finding, okay, where is it that we're teaching people how to use agents? Meg, do you remember how we use, let's see, using agents? I send the PR for these pages, isn't approved yet. It's waiting for PR approval. Ah, okay, got it. So we've got a page pending, but it's not really, it's not been merged yet. Yeah, not yet, but it's pending for review. Excellent, okay. Well, so it's answered, I guess. The problem is if we can't register every issue, we have no idea about the work we need to do, but it's okay. We're following with the sheet. So, yes, for me, I'm less worried about having an issue for everything we need to do than I am about having, cluttering things with incomplete issues that don't let someone start working on it with right away when they pick it up, when they begin their work with it. Okay, I got it. But I'm open, I'm open to other guidance. In this case, Oleg and I intentionally decided together, hey, we don't want to create issues that a writer can't start with and know where to, where to put the content. But it means a human being has to do the analysis that we just discussed. Okay. Okay. So there is another talk, I guess, maybe. Oh, we are already talking about the redirect strategy. Oh, right. Well, and let's, let's go further on that, right? So we, we are still using, we'll continue using vhost.conf redirects, but now you've identified in the tables things that still need redirects, right? We've got done needs redirect and these, yeah, these, the work is done, but the redirect is not functional, correct? That's correct. Maybe because they double slash and they because it needs redirects in fact. Yeah, good, good check. Okay. So there's exactly the starvation. Okay. So that one definitely needs the redirect. It really doesn't have it. So, so that's really good because and it's, we need, now how do we track or is it worth, you've, is the sheet enough to track it? We don't even need a separate issue. Let's just use the sheet. You've done the work here, Jonathan, right? To tell us. Yeah. This needs a redirect. Some issues, yeah, but we need, we have the issue link, but some of them I, I can't find, find the destination page. So for example, the executor service, I don't know the destiny. Oh, oh, now that's interesting. So we closed, closed the issue. Yeah. Ah, here we go. Okay. So it says 3384 resolves the issue. So maybe we need to visit the PR, check the destiny, and then trace the project. Yes. Right. I think that's, so in this case, the redirect will be to, executor starvation will be the redirect and it would need to, so let's actually, okay, once again, it would have been better if I had it in the issue. Yeah. That's correct. Now let's see if Google finds that for us. Executor starvation, wiki, wiki, wiki. Nope. So it's, so Google search isn't even, isn't even listing it in top results. Okay. Action. Use the sheet to submit our redirects, PRs for pages. Is that a reasonable action, Jonathan? Yes. Okay. I agree. Okay. And that one is not contingent on me. Anyone could do that, right? I can do it. I find the link. Okay. Great. All right. Jonathan, this is absolutely wonderful work you've done with the sheet. Anything else on, on the topic? No, just a question. Do you have idea, all pages in the sheet, our work we need to do about the migrations, or that there is another page from week to migrate to, because that link was sorted by visit. Okay. There is another one, or if we migrate all of the lines, we finish the migration work. Good question. So the sheet is assembled from 90 days of access data, sampled several months ago. Current access data may show more pages to, actually not may, will show more pages to consider. And let me get a link to the current access data, because that way it's embedded in this file. Let's see, and that is here, wiki-top-urls. Here we go, this one. Okay. So this is a summary of page access counts for the last 24 hours extracted from the Apache log files. And so this, and what I do is I have a script that runs every day, and grabs this, and keeps it locally on one of my computers. And so I just gathered at that time 90 days worth of these things, and use it to assemble that report. So, but this link is showing the last 24 hours, or what's the period? Last 24. Well, what it shows is the, shows yesterday's. Yesterday. So how can we have the last three months visit? I could share the data with you, wherever you'd like. I just, I have it copied on my hard drive. Let's let me confirm that it's still there. But the last nine days for now, you have this? Actually, no, I have, I have much more than the last 90 days now, because now I've, I've never deleted any of it. And therefore, I now have 180 days of data that I can, I can certainly share with anyone who would like to do any analysis of that. We just didn't want to keep this kind of data on that server. So I just use a little bit of this space on my computer. No, I guess. Well, it's because I guess it's important to know how big is the work we need to do. So if we can map in the, how they work, maybe you can plan and know how days we need to put on the work. Yeah. Yes. I think, well, I'm, I'm less concerned about any of the things that we need to do. Well, I'm, I'm less concerned about estimating the work and more concerned about being sure that any information that people find useful, we haven't, we have visited. Let's see. Just a minute. Let me look at this program just to see what it does. So oh yeah. Okay. So this, this little bit of Python just collects all the files, counts them. So here the top most accessed page is my reverse proxy setup is broken. And then counts go down from there and it goes down to this one. And how many lines does that have? It has 380 lines currently of pages that were accessed sometime during the last 180 days. Okay. Can you export to us? Is that the issue? True. Absolutely. Okay. Yeah. So, and let's see if I take out, we take out plugins because plugin conversion is a separate process, still 230. Oh, whoops. We should use a case in sensitive 220. Yeah. So how would it be helpful for you if I output, if I export this? Hey, CVS or or just the text file we can convert after, during the week. Well, or do you, would it be okay if I just gave you the raw data and the script that I run and you can modify the script as you wish? Because I don't know. They're all data it's enough. Yeah, it's only, it's only one and a half megabytes of data. So it's not actually that much data. It's each data file is only five kilobytes. And so, so they're not, they're not that large. How about if I will just put, because there's nothing confidential here. So I'm just going to put is, is tar acceptable or are you going to analyze this on a Windows computer? It's okay. You're on tar. I can start. Okay. So let me, I'm just going to create. Let's see. So let's go. I guess we don't need to send me file by file. Maybe that expression just the econ in the text file. Yeah. What I'm going to do is I'm just going to be lazy and I'm going to put, I'm going to put everything into the tar file and then I'm going to put it on a little web server that I run. So, and I will embed the link to that on Mark's HTTP server. Okay. Now if I click that and this, it should download it. It did. Okay. It's only 110 kilobytes. Okay. Thank you. Anything else on the wiki, wiki migration? Will you shoot me if I go off on a quick tangent? Oh, go ahead, Meg. Well, while you guys were talking this other stuff, I googled launch agent from console. The top two hits are from CloudBee support. And you go on down to wiki. I haven't reviewed the material carefully, but my Spidey sense tells me that we've got a problem in the doc there. That that just doesn't sound in it. Well, it looks like there's a wrinkle that it doesn't show up unless you've enabled the what used to be the JNLP block or Well, and that specific page you were referencing, Jonathan has a pull request to significantly improve the agent documentation on Jenkins.io. But Jonathan's pull request is blocked by the absence of time from a reviewer, i.e. me or Oleg or others. So, we just need to get the job done and review what he's proposed. I was looking at another one where you were going to delete two lines and just refer them to the regular Jenkins doc on how to launch. Yeah, right. And that was saying that that might not exist or there might not be anything there that's that good, but that's right. And Jonathan's Jonathan's actually created real documentation for us on on agents. And I think you used another page as the basis for that, right, Jonathan? It was there were two or three or four pages related to agent up to using agents with Jenkins. Yeah, I update the content with another outside Jenkins information. So I create new samples and I put their step by step how start the agent agent with using the pipeline. Yeah, there is a lot of good content there. Except to say that my Google search just confirmed that what you did needed to be done. So, right. Excellent. Okay. Anything else on the wiki transformation? No, I guess not for now. Thank you. All right. Excellent. Thank you. Thank you. This is wonderful progress, Jonathan. Thank you very much. I just wanted to ask Jonathan, are you covering both inbound and outbound agents or just just some some some agents on part of agents? No, I just I put there a sample how to use agents to help the main Jenkins installations work to execute the pipeline. But there is not so specific to a task or another kind. Just how to use a simple agent. I see. So we are not like doing any details about communication with the master or whatever it will be called. And agent? No, no, just the agent to use a SHH connection. Oh, so you're only talking about a SHH agent, I see. Yeah, it's just yeah, that issue. Okay. I guess it is one of many agents which can be enabled by Jenkins. Yeah, there is a lot of samples, but it's impossible cover all. Well, and once, well, thank you very much for doing that much already. I had an ongoing discussion with a user, this one right here, where Zocker 1.999 net provided a brilliant bug report on, hey, I don't understand how to use a Docker agent. And ultimately, the user said, look, I'm going to go back to using GitLab CI because I know how to do this. And it's good data from this user to help us understand how crucial it is that we describe agent use well. So your work on agents is a good step towards us eventually resolving this kind of problem for brand new users. Well, I'm not using Docker samples yet because I guess Vlad is the Docker, especially to do it. So I work on the basics and Vlad is the Vox guy. And I think that's that's we already need basics, right? We know we SSH agents aren't documented well enough. And there's lots to do just there. So yeah, that's great. Yeah. I mean, I've I've dreamed of some magnum opus that is everything you need to know about agents because there's all this I find no place to I find guidelines on how to choose whether you want a static agent or an ephemeral agent or, you know, why would I use Docker versus not use Docker. And I think the big reason for not using Docker is I don't understand Docker. But but I'm not sure there may be more but you know, just something that begins with a high level if I come to this, I know when I first started and I started asking questions about agents and everybody said, just don't touch them. We'll create a agent you need just, you know, now I'm seeing why. But yeah, and I just wanted to mention I can kiss if somebody needs to have write up why we need Docker, I can do just we need to define the place where you should go. Right. That's what I mean, the way we're going to get to the magnum opus that I fantasize about is a whole bunch of people contributing their little piece. And then we gradually put it together, right? Yeah, it's a great thing for crowdsourcing. Because I don't think there's anybody alive who knows at all. Yeah, it's from, for example, in my Brazil experience, several times, the people tell me they don't can't use anything on the Docker. It's not because the Docker, it's because the people don't know, for example, how to use janks in normal way. So they try to use the Docker, they can't. But the problem is not the Docker because you don't know Jenkins. If you can connect for any agent using SHHH by SHHH, you can use Docker too. So you need to grow step by step before go to Docker. Right. And I have a hunch that a lot of people are not using agents as well as they might because they don't understand it. If they get something, like your guy, he's going back to GitLab because he knows how to use that. They get something that works and that is their hammer for all nails. So. Exactly. There is no magic. You need to study. Right. Good. Anything else on agents before we go to next topic? Okay, next topic. So Vlad, do you want to be the voice to describe what we've learned so far on the Windows upgrade guide? Oh, yes. Absolutely. I would be glad. I learned personally how to do blogs and just simple, by simple referring to the blog which was done before by Alex or I guess on the same topic on Windows upgrade. But it is just beneficial for me. I'm not sure it was beneficial to you as the help on it, but I saw that. Also, I learned that there is, well, we need to differentiate between Windows installer, which is kind of outside of Jenkins. Or external upgrade and internal upgrade, where we're upgrading configuration and versions inside Jenkins without doing install. So there are two approaches to upgrades. One is internal. Another is external, which is nice. I still found some issues which I referred to these in the comment for instance, it is through internal upgrade. It's not possible, for instance, to switch to any version of JDK, which we or customers may be interested in. For example, above 9.0 version of Java. For instance, Java 11 is not accessible using internal upgrade. And I'm talking about upgrade from 2.2.35.2, which is 32-bit version running on Windows. Also very interesting issue, I noticed that before when doing this upgrade, before the weekend at least, I was able to get my Windows installer to 2.35.3 running on Windows. I'm talking about MCI executable, but today, for instance, and I think over the weekend, I was not able to do this because of limited, not enough privileges to run Jenkins script. Although I switched myself to administrator, changed properties and run as administrator on my Windows box, I thought that maybe it was because I was user before, but even by switching to administrator, I'm still not able to run Jenkins script. Well, some issues I'm just sharing. Also, I just wanted to verify when we're saying 2.2.35.4, I guess we're referring not to LTS release, but to weekly build. Is it correct or maybe? Actually, that one is LTS 2.2.35.4, so weekly has been using this since it's 2.232, April 2020. But the first time we did it for LTS was actually 2.235.3. And this is what is referring right now, download page. So now your permission issue is one that needs description in the upgrade guide. I think I know how to solve it and I'll send you a comment. I'll include a description in the upgrade guide and a link to an article and then we'll have you test it just to be sure it works. Okay, and we're talking about 2.2.35.4, not 3. Well, so this is a complicated one. 2.235.3 did not initially deliver the new installer. But did deliver a few days later. We had Windows infrastructure problems because in the most recent episode, Microsoft inadvertently blocked us from building by releasing a surprise change in their virus scanner, which said one of their own tools had a virus. And they fixed it very quickly there afterwards. It didn't take long for them to fix that, right? Lots of people told them this is a problem. They immediately fixed it, but it just happened that it blocked our ability to do a build on that specific day. I guess it was not coronavirus that is why they fixed it. Yes, this in fact was much slower to mutate than coronavirus is. So yes, absolutely. Mark, I'm curious is your solution related to the overall permissions in Jenkins or is it a Windows thing? It's a Windows thing. There's a there and it's the nature of Windows, right, that they correctly don't grant permission to login as a service to every account and they shouldn't, right? Because you and I as human beings should not login as services. That'd be a way if somebody compromised my password and then as login as service permission now my computer is has become their bot. So don't allow you to log in as a service is a good thing. So you need to log in before starting or like doing anything with the service I told you. Well, it's that there's a there's actually a setting in Windows that checks that now we probably ought to now that you mentioned it Vlad we ought to need to submit a an issue report that the validation check in the installer should confirm that the user has login as service permission. Because right now it's checking that the password is valid and that the user exists. There's no reason it can't also check that the user that's been provided as a service account can actually log in as a service. Vlad, you willing to take that action item? Well, if you want to take, yeah, but I would be glad to help Mark on that. The item is to file the issue not to fix it, right? Right. Yeah, so well, and really it's what I would propose is don't submit the issue until you verify that the new instructions I give resolve the issue. Because the new instructions I create will be the proof that yes, in fact, that was what the problem was. And until and then then you say, Hey, we need to check for this instead of just documenting it. Yeah, sure. So I will not submit the issue before I know how to fix it. Right, exactly. Okay, good. It's related to login as service. All right. And do you are talking about this login as service inside installer or just some some stuff needs to be done in general to assign privileges? So it's a this is a you'll have to open a with a Windows control panel and navigate to a specific section in the permissions detail in the permission section of the Windows control panel. And you have to open up a property panel, add a new user, etc. It's it feels there's a great page that I read to to guide me on step by step, which things to click in the in the Windows UI. It may be it will may become part of administration guide for Windows folks later in years. Well, at least it needs to be part of the upgrade guide for sure. Good question. Should the should the Windows install instructions? I haven't checked to see now you've now you've prompted something. Okay, Jenkins Jenkins.io. What do the install instructions currently say? Installing Jenkins, we certainly have a big section on Windows. Yes, here is our here are our instructions. It's and and that's that's probably reasonable, right? That's not unreasonable to say download it and run it because but this might be a good excuse for us to write instructions here, which use what you wrote and what Alex Earl wrote in that original blog post, right? This this section on Windows should probably include those screenshots from the installer. Yeah, and also I was thinking about maybe in case Governors Board will approve maybe these instructions like this part of administration related to administration services may become or may be included inside administration section allow documentation because we'll have Jenkins. Yeah, that that one I'll need to think about. I'm not sure where we currently describe port numbers and port access. You know, how do we tell where do we tell people oh your Linux computer must allow the port that you're putting Jenkins on whatever it is your Windows computer same way. It's a good question. There is another question about the Windows 2. For example, usually here in Brazil we start Jenkins on Windows Server, not in our common installation. So maybe we should create some documentation we need to know how to create on the Windows Server 2, for example. I think it's the same. It's a good question, but I think I just remember it's the same because the because it's the server installation, there is different policies and ports and ways to running services. Ah, I see. Okay, good. It's not the same the common common user Windows. Okay, good question. And so and certainly there are we know there's one example where most Windows Server machines probably that are running Jenkins are probably also running at a copy of Internet Information Server, right? They're probably serving web pages somehow. And that how do you configure those two so that with Jenkins and then there's SSL on Windows with Jenkins. Those are topics that haven't been touched yet, and certainly users would love to know how to do that. And just a question mark. Do contributors like us need access to Windows Server? Run a Windows Server operating system, for instance, or it's not necessary? It's not necessary. I don't think so. I can start up CloudBees funds me with a Google Cloud platform account, and I will do so. I can do some double checking safety checks to confirm that the Windows Server instructions that the Windows instructions work for server as well as desktop. I've done all my checks on multiple copies of Windows 10. But it's not hard for me to spin up a Windows Server, run some tests, and then discard it. All right. Anything else? We've hit our hour. Thanks very much. Thank you very much, Mark. All right. Thank you guys. Thanks. Recording will be posted later. Thanks a bunch. Thank you. Bye everyone. Ciao.