 Welcome, everyone. This is the Jenkins governance meeting. It's June the 12th 2023. We've got a number of topics, including news, several open action items and Jira upgrade and budget and expenses and community activities summary that I assembled. Are there any other topics that people would like to be sure we add to the agenda. Okay, then let's go over the topics we've got. So by way of announcement, Jenkins 2.401.2 release candidate will be available this Wednesday special thanks to Chris Stern for his work as release lead. Debian 12 has released last Saturday. It's a nice operating system upgrade. Important to note that it does not offer Java 11. So our documentation will be updated to guide people to use Java 17. That way we don't have to explain that on Debian 12 Java 11 is not even available from the operating system provider. And the last topic I had was a reminder that the CDF technical oversight committee election is in progress. I've been nominated from the Jenkins project. There are only four seats on the committee. Six candidates I would love to have your vote. If you've received an invitation to vote. Any questions on any of the news items. Okay next topic then is the action items. So let's take a look here. Easy CLA documentation not happened yet. I believe there was some interaction there that might need more on this. Do I remember correctly or am I am I forget am I off base. On the documentation part. I'm not too sure but I think last week or the week before Chris Stern submitted an ICLA. Ah, that's what it was. Thank you for the reminder. Okay, so they submitted the sign PDF and one through the Linux foundation. He took both paths. Yeah. Very good. Okay. Thank you. All right. And that was now that I think about it I remember that was so that Chris wants to join the Linux, the, the security team. And is interested in that if I understand correctly. But I confirm that Chris isn't in need of ICLA now that it's just for future action I think. Okay, so the ICLA was not strictly required. No, for current at least not yet. Great. All right. Excellent. Thank you. The topic was a poll request to come combine sub project and special intergroups in interest groups into a single concept. I have not done that yet. I've gotten started on other things but not not successful yet. Retire the Jenkins the Chinese Jenkins site is in progress in that it's no longer on the header. But we've got more to do in terms of getting the sub pages under the site redirected right now. For instance the install instructions are two years out of date. And we've had an awful lot of changes installation instructions in two years. It's really bad that we're showing Chinese users incorrect and flawed install documents. I'm going to go on Kevin shoulders Kevin with trust you'll take it forward. Then I had the item to archive the governance meeting notes sorry no progress. Retrospective on the signing certificate renewal progress I've opened the document but not started that again sorry, I am proud to say the next one I have completed the reimbursement process for the code signing certificate is done I have the money. The process worked. Thanks to Oleg especially because it took a number of steps for us to figure out how to get that money out of the account through the Linux foundation into my account. And we have one other reimbursement that's been successful. Thanks to Vodic Filonia. So, positive on both. Any questions on the open action items. So the next topic is the Linux foundation ticket to upgrade from Jira eight to Jira nine thanks to Alex for this one. He noted that Jira eight is end of life within the next. What is it Alex I think it's within a matter of a few months or less than a year. Okay. Yeah, in October of this year I guess. So, I had asked the question, Johnson when from Linux foundation asked me on 30th of May if we had a specific timeline. My answer was not really so long as we're done with Jira eight well before the end of supportive of Jira eight. We're fine with that. Anyone have any concerns about that do you want to do want me to put pressure on them to say no we need to do it much earlier than say August. Okay, great. So the next topic then was budget and expenses, and I'm proud to say that Vodic Filonia has been reimbursed. Now, the rest of you may not say that's a very big deal the $52 and 99 cent reimbursement but it's taken us as much as two years to get this reimbursement finally closed he confirmed the money has arrived in his account finally. It's a reminder that when we have expenses the easiest way to do it is have somebody from Linux foundation use their card to pay the thing, rather than any of us using our own funds and trying to get reimbursed it's just much easier if we make them do the payment. They'll figure out how to route the money then. Any questions on that one. Okay, next topics then our community topics. And the topics that I've got, I'm open to any of these that that are particularly of interest to those of you who are in the meeting if you if you want to say, Hey, let's talk about this one first or that one. I'm happy to move things around. If we change then we've got an active project with JFrog, trying to reduce the bandwidth that we're using on repo dot Jenkins CI dot org. We're proud to say that we reduced by 20 terabytes a month in April compared to previous months. Nice reduction, but they really think that we need to do more. And we're, we tried, we did a test drive brown out a reduced functionality test by shutting off the J get copy that we have, and it was disruptive. And because of that we're going to have to now talk to JFrog about what other alternatives are there. So we'll, we'll be doing that work with the infra team to identify ways to reduce bandwidth use. The most flagrant abuser of the bandwidth has already been banned, but they're confident there are other things we can do to reduce bandwidth use further. And since they're paying for the bandwidth. It's important to them that we find ways to reduce it. There are some questions there. Okay, prototype JS. So this one is a special thanks to Basel Crow, and to Tim Jacome prototype is a JavaScript library from 2010. Last released I believe in 2013 or maybe 2015. So in the world of JavaScript it's ancient. And it disrupts the use of JavaScript in other ways in Jenkins core. So the progress is being made to remove it completely from Jenkins core and from the entire plugin suite. It's, it's references to it have been removed from Jenkins core already as of 2.406. Key plugins have received pull requests. And there's a feature flag so that those of us who are testing and interacting after 2.406 we can actually disable prototype completely to see how it behaves. So, Basel is there anything additional you wanted to share on this one. No, I think you covered it. So it would be great to see more help to adapt plugins, because that's really what we're blocked on in order to complete the project. Right and this this sheet that Basel has been maintaining is a is a marvelous piece of work in terms of seeing where can I help. So when when volunteers come saying hey I'd like to help the Jenkins project. Any one of these rows is a good way to say you here you can learn some JavaScript and help the Jenkins project in a substantial way by applying to your work to one of these plugins the red lines are very much wide open ready to have somebody help. Yeah, it was 22 to 24 are a good place to start because they're a high installation count, you know in the 66,000 installs range. So I don't think we could realistically do anything until those have been adapted, for example, agreed. Thank you. Any questions to Basel on the progress on prototype JS removal. I have a question so what is the plan when do we want to switch it in the opposite way so it is enabled or disabled by default and users can enable it if something's wrong in their machine. So I don't have any poor quest to do that so I, there's no plan on my side. Think Tim has a draft poor quest to do that. And I don't know what his plans are for it. But you don't know if it's, if it will be part of the next LTS release or not yet. Okay. I'd be shocked if oh go ahead Basel, I shouldn't answer. No, I don't have any, I don't have any knowledge of whether when Tim plans to take that poor quest out of draft state, but until the spreadsheet is more green. I don't think we could do it anytime soon. Yeah, that was that was my assessment because if we, if we dropped prototype today, any user of the credentials plug in and that's 280,000 users to it would be broken. Right. So, so that's the today's not the time and I think the same thing applies we've seen at least down to the 30,000 level of the 10,000 installs level that we really want to get those resolved. Okay, next topic then is HTML unit three upgrades. So this one, the HTML unit project, change their package name and change some of their API's and their upgrade from two to three. And Tim Jacome and Basel have have done the preparatory work. Tim has tooling that creates poll requests you can see the hyperlinks here. The first harness has been upgraded core has been upgraded. The plug in palm has the plug in bill of materials, and there are 150 open plug in pull requests. So there's, there's still a lot of work to do there, but the work is proceeding. It's this one because it's all test infrastructure is not a risk of breaking Jenkins users at runtime. So the challenge is if you choose to upgrade to plug in palm 4.66 or newer, you need to adapt yourself to HTML unit three. Now Uli and Alex I believe you're both also plug in maintainers are there any things that you've experienced in your plugins around this that you wanted to highlight. I am not using HTML unit. Oh, good. I'm using real user interface tests with acceptance test terms. My plug in sit. Yeah, it seemed that HTML unit is something one should not use anymore. So that's, I completely removed it in from my plug in tests. And I'm using real browser tests with 80, 80 h. Great. Thank you, Alex anything from you in terms of your use or are you use doing any work in areas that use HTML unit three. I think I have received one PR from Tim updating and it went pretty smooth updating the imports and changing a couple of methods but that was basically it checks when green and I merged that last week. Great. Excellent. Thank you. Any others who want to highlight any items. Next topic then is the guava 32 upgrades surprise and success story Basel you want to tell it a little bit. Well, the upgrade of guava in 2.406 had a bug for Windows users when creating temporary files. And there were two plugins that used guava to create temporary files, artifactory and checkmarks. So those were broken on Windows with 2.406. We're able to get in contact with the guava developers, and they fixed this problem in a new version of guava, which we should be releasing in Jenkins core tomorrow in the latest weekly. So that should resolve the Windows issue that we saw. That's it. Thank you. Thanks very much. For me that's a great success story that one open source project helped another detected a problem that other project provided a fixed promptly and we got it integrated. Very nice. And they added a Windows CI build to guava so they shouldn't have any problems like this in the future either. And that's that's especially valuable I think that yes that means they're testing Windows more than they ever tested it before. We like that. Great. Google Summer of Code is continuing. The end of life for Centos seven has been announced, and we're going to accelerate the visibility of that warning so that it will begin appearing to LTS users at the end of this month with 2.401.2. It's been approved by Tim Jekyll and merged so thanks very much Alex for your inputs thanks for Tim's and thanks for bozzles. Last item or next item was on GitHub Enterprise and this one Alex maybe you want to describe what prompted it you gave us a blog post about it. Tell us more about how that's that's working etc. Actually, I had a chat with Gavin McDonald from the Apache Software Foundation, a bit earlier, and I was checking something with him. And he informed me that the Apache Software Foundation GitHub organization is on some special plan that is not available for regular people. So, I went ahead and ask it up if they could apply that to our organization too. And with the help of Kim Jack, Tim Jacob approving it as an orc admin, they applied that to our organization. That brings us a certain set of enterprise features. But most notably, as you probably have seen already and core these Jenkins dash links directly linking to our Jira instance. Auto link references is the proper word for that. Let us opt in for everyone hosting the plug and within the Jenkins AI organization and given our heavier lines on Jira. This is probably one of the go to features for us. Thank you edit the. Oh, go ahead, Mark. No, no, excuse my interrupting. All right, you edit the task list beta to the enterprise part that's actually not part of that's not part of the enterprise part. Okay, so it's just a GitHub feature that we're using. Just a private private beta I enrolled the organization in. Got it. Thank you. Okay. I think Tim did that because sometimes they want approval from an orc admin sometimes they don't I don't know so. But we have it. All right. Any questions to Alex on the new new capabilities from GitHub. Okay, last item on that I have on the agenda was related to launchable. And Basel, do you want to say something about launchable you prefer me what how would you like it. Well we started using it in Jenkins core to on the windows. See I builds. So they, they're a lot shorter now for pull requests because we're running a subset of windows tests in Jenkins core. And so that was the first of the next steps in the next steps bullet a few lines down so that's been done. And the next step after what has been done already is the second bullet and the next step section which is to add subsetting to a th. So I'm hoping to get to that sometime this week or next. Thank you. Thanks. Have we had any cases where we've detected a problem that somehow was missed due to the subsetting that we're doing on windows. I mean, if we if they're, we're still running full builds on builds to the main branch. So if there was a test that was failing in a PR. But we didn't but that subset it away we'd notice it as soon as that PR was merged to the main branch. We haven't noticed anything like that. So pull requests run a subset but the after merge to the main line, it's runs the full set. Yeah, okay. And the only thing I can think of or would have caught problems would be that guava issue and I've caught that in PCT, but we don't run PCT on windows at all. So we didn't catch it there. I did run PCT with the guava upgrade but it was all on Linux. Right. So, I don't think launchable would have helped us in that case. Thank you. Any questions to Basel on the launchable experiments. All right, that concludes the topics any other topics we need to discuss today. Thanks everybody.