Jenkins June 2021 Contributor Summit, 25 Jun 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: State of the Union

Description

Jenkins Contributor Summit June 25, 2021 State of the Union

Status updates from the Jenkins governance board and Jenkins officers.

Slides at https://docs.google.com/presentation/d/19k3exBMmR7hWWXdsGFyGPQZoSHDKYIgZEYzc1khBLt0/edit?usp=sharing

A

Hello, uh welcome to the jenkins contributor summit today is june 25th. So thanks to everyone who attended the cdcon, uh if you attended the full conference, you already had three days full of various talks, including many jenkins talks from our users from practitioners from tech leaders, and today we actually focus on contributors, and we will talk about how we evolve jenkins. We will talk about the various issues and topics we would like to address, including interoperability, technical changes and the junkies core.

A

We will also have say a trek for newcomers and I think we can start.

A

Okay, so it's switching now so what we will have uh today, so we have multiple sessions. I will show the full legend later today. We will just have overview of what is going on in jenkins. We will do some announcements, we talk about project and team updates and then all our sub projects, special interest groups, essentially we'll be also able to make the updates uh the idea that this session will be related to the shot.

A

If you want uh here's a link to the contributor simon slide deck and the problem mark could shade in the chat so that everyone can access it. So our slightly deck is already.

B

I'll take care of that. Oh, like I'm, I'm working now I don't know okay, so.

A

Yeah uh welcome to the jinx contributor summit. This is if this is a first time for you. I will quickly explain what happens. The idea is to actually gather contributors from around the world, at least from a time zone. We have so this time it happens in the americas time zone, and I hope that the next time it will happen in a bad time zone so that you will have a different uh slice of the community participating, but still we have many people and hopefully we'll have more people joining what is the contributor.

A

So this is an event which is focused again on contributors, uh it's hosted by contributors, so you can see various uh not so finished slides. You can see some organizational blunders, etc. So we are not doing a marketing conference. We are focused on the contributors on this live discussion. We invite everyone to participate to show live demos. We will have a session for uh lightning talks which is still available. We will have end user panel and everything is about conversation.

A

So the goal of this summit is not to show many cool presentations, but just to actually achieve something as a coming. It's a one-day event uh actually event will last for six hours and uh yeah. If legend is available uh on the links, so once you get the slider links, you can just navigate, because this sliding is public.

A

Okay, there are a few items for us, so firstly, if you're a newcomer, we have a newcomers track where mark will be talking about how to contribute to jenkins uh this time. This is the first time when we have end user panel. We have multiple users of jenkins joining and they will share their experiences. uh Talk about the issues that they experience with junkies and expectations from its evolution.

A

So this is the first experiment and I hope it will be successful and if so, let's keep doing these sessions and then we will have a pretty long unconference, uh where we will talk about topics which we discussed before the summit in the community.

A

So there was google doc where everyone was able to contribute, and we agreed that we will have three parallel sessions. So what are the topics here? I take some interoperability jenkins, sex, github actions and jenkins cloud integrations, open, telemetry and observability.

A

All of these topics related to other projects, uh so it's basically interpretability integrations, and this is how jenkins works is really about integrations with other systems, uh parts of the continuously foundation or not. uh We are happy to support these integrations anyway. Also, we will have a discussion about uh court jenkins and its evolution, so there is a topic about the plug-in end-of-life policy.

A

We will talk about the goal of ability and terminology update in the jenkins core, and then we will talk about java, 11 and 17., and last but not least, we have a discussion about jinx's kubernetes separator, which is a new sub project to join jenkins, and if you use jinx and kubernetes, this is definitely something you would like to consider after that, we'll have ignite talks and demos.

A

So if you're a member of the charity, if you have access to the agenda, google doc feel free to make proposals, because we still have a lot of slogans for unite talks and demos, and if you want to present anything about jenkins uh v, slides without slides, just live demo. uh Please do so. This is the session, especially for that and then you will have a closing session and a happy hour which is likely to happen either zoom or in a work adventure depending on whether we will get the system running by this time.

A

Okay, so you may have experienced some issues with connecting uh to this event. um We will be sending links in the chat later and this uh temporary links for all these sessions, and I believe that mark will be able to confirm these links going forward and yeah. In the worst case, we have a slug I'll share the links where we can do the destination, so you may ask where to go for newcomers.

A

Yes, you start from the newcomer track and then, if you feel that you're good enough information, you can just switch the classic contributing to change track and it's up to you for seasoned contributors. Of course, we recommend to start from end user panel, because this is a session when end users will be talking to us yeah, probably they will have some uh topics to share, maybe not that positive feedback, but this is why we invited them so- and this will be in breakfast session.

A

So usually it's uh contributors, presentations and users asking questions but we'll uh try the opposite and then of course break out sessions, and everyone is welcome to join our happy hour. So we will have more chat. If you don't fall asleep by this time, it will be something like um 8 p.m, utc I believe, or 7 p.m, so not that late.

A

Okay, now, regarding discussions again, everyone is welcome to participate, share your feedback share your proposals and do liking talks. We also have a code of conduct. This code of conduct is based on contributor covenant. So please be nice respect everyone, and there is a lot more words, but that's like that. For the conversation we have a uh jenkins contributor summit channel on the cdf slack.

A

If you want to join uh there is uh there are guidelines how to join so you should follow these guidance so that you can can connect with arbitrary email and it's basically just uh clicking this link entering your email receiving the invitation and then accepting it. So if you want to join, please follow that and you will be able to participate in the summit, so we will use this slack channel for all kinds of communications and again we will post links to the slides, etc in parallel with the chat.

A

Okay, this is all with the introduction. Now we have state of the union. So this is a summary of everything going on in jenkins, and this time we made some improvements, because one of the feedback topics from the previous years was the state of the union is too long.

A

We addressed this feedback, and now we will try to do all the updates in 40 minutes, so we will have something like 10 people who will be doing a lot of ignite talks right now and gets ready. So, let's see how it looks.

A

Okay, so I'll start from some governance updates and then we will switch to project updates, etc. First of all, I would like to thank markie jackson, uh so when the last time he was a board member and jenkins events officer, marquis has decided to step down, but yeah thanks a lot. uh Follow your contributions, marky and when you're ready uh welcome back to our community, then yeah. We will talk about evelyn and me later. We will have some updates on communication channels thanks to our contributors, treasury and community awards.

A

There are many other topics we can discuss during breakout sessions and later so yeah since marquee stepped down. We invoked our process for temporary board members and we would like to congratulate evelina wilkers for joining the jenkins government's board. Effective, I believe february 20th, so evelina will be a fifth member of the uh jenkins governor's board, the one who is me kevin, morgan, uh willie, hoffner and kosike, and everyone intends focus on end users, communications and, for example, the today's end user panel is actually inspired by evelina, unfortunately she's unable to join today.

A

But we follow up on the discussions we had for the last month. uh Okay, also update for me uh yeah, I became uh uh acting changing servants officer, so I will be. uh I go to contact if you want to organize an event regarding jenkins, etc, and I'm also joining the cdf for over technical oversight committee, starting from july 1st. So before that it was joseki representing the jenkins project there and if you have any questions to the cdf, just contact me and we will be able to follow up and again my job.

A

There is representing the junkies project, uh not making decisions for the junkies project, as I announced in the mailing list, and any feedback and suggestions will be always welcome.

A

Okay, a few major updates on the communication channels. I believe damien will have uh some more to say later or maybe damien would like to just summarize the updates. Now you like to do it. Maybe okay, then it's me so: firstly, rc used to be our main communication channel. It's not long. Actually, true, we still use rc, but it's just the first class citizen along with many other channels, uh but we had to be created from a free note, ifc to liberate child during due to the events.

A

Well, everybody heard of in the open source community and now for our main channels on the library chat and we cost about eight other channels, because we decided that mailing is a guitar enough for this purpose. So if you haven't migrated to liberation, yet please do so because jenkins and also uh hosting um and the release teams, are there and also infrastructure teams. So if you participate in various infrastructure related activities, it's definitely a good channel to join.

A

Regarding the rest, yeah, you may have seen these slides. We actually have too many communication channels and thanks to kevin morgan, olivier and other contributors, you actually handling this pro and we introduced another channel to dissolve fishes with too many channels. So, thanks to this course, we were able to deploy a community jenkins io. It's a sponsored service where we intend to move the most of congress stations. It's not that we abandoned millionaires. We will keep using them, but many conversations for special interest groups, et cetera, can now happen in the community jenkins.

A

I o and you to create today's presentations. We will be referring to it multiple times. So, if you want to have a discussion on any topic, it's an additional channel. You can consider and yeah. If you have another in this course, it's a kind of forum where everyone can contribute and create their own topics and the what we have there. We have uh placeholders for using jenkins for contributing for various community things. Also language groups like espanol jenkins, jenkins rule, and they will also change jenkins.

A

Soon and again, everyone is welcome to contribute and expand this platform. As a central communication point for unofficial junkies communications, official communications will still remain in the mailing list for the time being,.

A

Okay, um worrying update about our cash flows. So, as usual, we have not that much updates and we don't have much money in our offers, but we have some uh so we have around 30 000 dollars over the past year. We invested a lot in outreach. We invested a lot in our infrastructure, so we started from 30k in 2019.

A

Now we are down to 13.. The most of them are still allocated on sbi, but we will be moving to lfx crowdfunding soon it's arranged. We just need some uh governance of invites then we'll do the migration and- and currently we have quite a lot of money reserved for google summer of code, so it's for google summer, of course, 2020 and 2019, because we've been able to spend all the budget because of coronavares etc.

A

So we have some money reserved and we will still receive the event for this year so that we will be able to sponsor students and private travel plans. Should the situation normalize by the by this autumn and generally of all contributors, please reach out if you have ideas how to spend money. This is why we receive money from damage donations. We expect to spend them at a facilitated agency.

A

We have multiple initiatives on our roadmap and we invite everyone to invest so that we could actually spend them. For example, documentation projects, design projects suggest developing new features. All is possible and yeah. Thanks to our sponsors, um this list is not complete, so yeah the djing, this project uh so kind of survival in this amount of money. So you can see 13, 000 and well it's not that much.

A

We spent a lot more on infrastructure every month and thanks to our sponsors, uh who make it possible, continues delivery foundation, aws cloud, bs, red hat, github jfrog, also also all of them invest a lot of money to support jenkins via various programs and also there are many other companies which support us through open source programs, uh for example uh yeah. I just referred to this course. It's sponsored by.

A

Dispos house this course construction kit. They incorporated basically the company between this course and there are many other sponsor companies helping us, bring our infrastructure and other services and yeah thanks to everyone who supports the project.

A

Okay, uh last update before we switch to more interesting parts. um The cadiz liberty foundation uh community awards. So if you participated in the ceremony yesterday, there were multiple awards, actually eight hours plus project hours, so maybe uh 14 in total uh uh focus on different uh contributors, and today we are multiple uh jenkins awards because is created project. So we were able to have three hours this year and most available jenkins contributed mostly to jenks and security mvp, and we would like to congratulate the winners of these awards.

A

So tim jacom second year in a row is a most valuable agency contributor and it's well deserved. Thanks for your contributions, team to various areas, including infrastructure and user experience, congratulations, code, etc, etc. It's hard to say where tim doesn't contribute. This year, danielle yeah daniel has been a security officer since 2015.

A

When this position was introduced, and this year we were finally able to vote for daniel, because this year everyone can be nominated before the club is employees apart from receiving ours, but at this time yeah. Finally, we are able to recognize daniel who has been driving security and the most of the advisories over the past years and well, it's a serious job.

A

I believe we had almost 100 advisories by this point and a lot of security issues disclosed all of them requiring donation and again thanks daniel for that uh so yeah, the most valuable jinx advocate. That's me uh I can share updates later I won't raise myself and what I would like to say that cara de la mark, currently the leader of the agency's cloud native physique, she became top guitars evangelist because jakarta is one of the top promoters of top and also for secrets, management and githubs. So yeah congrats, carla.

A

And I guess that's it for today and we can start from uh project and teams updates. These updates will be relatively short because we had a contributor summit three months ago, so we will have reduced updates compared to previous events, when we had six or nine months updates. Still, we have quite a lot of content and we have something like 30 minutes left. So let's get started security.

B

C

I think you may need to promote vadak as a as a panelist. I don't seem to have the permission to promote him. So sorry for the disruption there. Okay.

A

Yeah by the way yeah, I have a slack open on my phone. So if there is any destruction you can just pinch me there. Oh thank.

B

D

Okay, so I added uh why dick anyone else issued that.

E

So could you hear me yes.

C

Thank you go ahead. Please.

E

Thank you alex for the reaction. I didn't thought you were talking so much until the the point there so perfect.

E

uh Next, just for the introduction I'm talking here instead of uh danielle back daniel, is just taking some pto for having a longer weekend so.

E

E

So to reply to your question, danny in term of a number of advisories since the beginning of the project, it's 124., so not everything from daniel alone, we were a team recently about security. At the beginning, it was mainly koshuk doing the thing so just for the information.

E

In the last three months, we published 10 advisories with a total of 46 vulnerabilities in total. So that's interesting, the number of growing and things like that. So we need to keep the pace in a sense. Please next slide, so uh what we did in particular outside of the advisory directly.

E

Recently we did some mentoring for a master project in the jenkins security area. It was with four students from marseille in the source of france. They were doing some auditing for a lot of plug-in, perhaps as a maintainer, you have received some of the report from them or things like that. It was during two months and at the moment 14 vulnerabilities were published, so we were very interested in the the outcome there, but also in the investment during the process during the project. They were very curious.

E

All the things there were interesting for us for them for the project, so that was just a very nice success. From my point of view, just in case there is a blog post. If you want to read it for for more detail, and especially at the beginner, the bottom of the blog post, there is a sentence saying that, if you're interested in such project as an intern as a student, as just someone wanted to do a bit more security things, please contact us. It will pleasure that we will set up something to work it together.

E

Please next slide sorry, so also concerning the different initiative we are doing recently, we got another xss in a credential, so that was impacting a lot of instances, and so we decided to re-want to be together to be the effort we were doing with content security policy innocence with that epic, that we have the link at the middle of the slide. It's something that is really interesting in terms of protection against xss, especially in the context of jenkins.

E

We have a lot of javascript that are inlined inside the different page, and if we are removing them, we will unlock the possibility to have a csp configured at the instance level and so protecting against a lot of excesses, good feeling at the moment, 90 percent. Something like that. So if you want to help on that arrow, please contact us as well. There is information in the ticket and also you can see some pull requests, sometimes about unaligning javascript content. So the topic is interesting not only for jenkins.

E

If you want to learn more about security in general, that's also a good opportunity. Next slide. Please and also alex r was doing a very good job with the hosting request, because he was doing some preliminary tests preliminary. What did in terms of security. It was mainly about like checking if there was some credential usage, some method that could be dangerous and things like that. This kind of preliminary audit was very useful for us in the jenkins security.

E

Do not spend time to do full audit on every hosting request, and with that quick approach that was very useful to save us. A lot of time because alex is doing the auditing, or at least was doing the editing during the hosting review about the quality and things like that. So if there is any volunteer about this kind of part, this kind of job, please also reach out to us. That could be useful and I think that's quite done for the things we did.

E

What is next for the security team, so soonish we will have a new advisory for the next core release. With some vulnerability. I cannot disclose anything at the moment. You will receive normally information soonish and we continue to do the different auditing. We are doing for the different plugins and things like that so expect to robot us. So in the future, I would say we will find something else. Thank you.

D

Thank you verdict. I will do a quick update about other topics related to junkies, secure software delivery.

A

So this is a follow-up from the previous contributor summit. When we discussed this topic uh in depth, uh one of our follow-ups was to continue with job 229 and he's developing his components and I'm happy to say that the this initiative got some progress.

A

It got some adoption and actually you can try it out if you maintain junkies plugins. So it's a fully available fold option. There are some things in there. For example, there is an energy stock project related to improved versioning of releases in our city flow. But anyway, here you can start using that and you can start adopting new plugins if you want to deliver jenkins against uh for every pull request and probably merge commit okay, another update related to security.

A

uh We started the joint partnership with the linux foundation team, uh particularly lfx, and we sneak uh and uh the idea they that lfx security is one of the uh hosting host services offered by the linux foundation, which intends to perform all kind security. So currently it has two services inside one is sneak and it's focused on dependency scanning analysis or some license scanning and compliance and soon there will be blue bracket introduced.

A

It's already available for preview for limited number of projects, so it will be basically analysis for secrets in the code base for misconfigurations if you use configuration files and those for uh common mistakes in the code uh we participate in the project in order to adopt the new version and, if you're interested to discuss, why uh feel free to reach out. So there are a few examples of how it looks like so this, for example, dependency analysis for jenkins, with vulnerabilities being disclosed. uh We also can get the issue details with references to cvs.

A

um uh Junk is gonna issue its own cvs, because we are cna thanks to daniel beck. So for all our advisors. Success will be released, information about uh variabilities on our own and they become available in the database, and then they become available in our scanners and same for license analysis. So you can analyze what licenses have been used and sometimes components use licenses which you're unable to use inside your company, and you can analyze that. So I'm showing you a red herring case when we have hpl3 library configuration is called plugin.

A

Please be sure it's not a concern in this case, but it's nice example that we definitely need to work on false positives.

A

Okay and yeah, as I said, we participate in a pilot project. So jenkins was one of the project. First projects to join and there will there is a lot of executive zero. It will be adopted by the jinx infrastructure team and it will be adopted uh for jenkins once there are particular features available.

A

We are waiting for them because currently we cannot really adopt because of massive number of false positives, and once the fix is available, we will be happy to try them out and to stop our skying pipelines powered by sneak and elephant security. So thanks a lot to all contributors who are looking into this story.

A

Next topic is jenkins infrastructure. So damien, would you like to check over.

F

Yes, can you hear me is the sound okay? Okay, so I'm replacing olivier today he has a child who is sick, so I'm replacing you he's excused and a good excuse. So can you switch to the next slide? So the what happened during the past month on the infrastructure, um the cost increased? These are facts. uh It's almost 2 000 bucks per month. More you see that also the the difference be the the repartition between aws, azure and other sponsor on the cost. We can evaluate as shifted a bit.

F

We depend a bit less on azure next slide, please. So we are working on trying to determine if this costs increase come from things we can work on or if it's intrinsic, so we have more than loads of more plugins more people visiting. So we cannot say stop. This is an intrinsic intrinsic cost increase, but there are still some levels. We can work on to be sure that we control the costs.

F

So we'll speak about that, but mainly we have two main personnel that are responsible for the services. We provide the users that download, plugins or visit the pages, and the developer will use ci chen kim scio in order to build the plugins to contribute to validator codes. So these are the two uh kind of person that we want to serve next slide please.

F

So one of the most important achievement has been the download mirror it's for the person of the jenkins user. So what has been done is providing more visibility for people and better performances and will reduce the time between one artifact, a plugin. A core release is published and deployed and the time it's available for everyone across the world, one of the main services that you can use as a jenkins user.

F

If you have any issue or have question, you can add on the query string one of the three elements: that's on that example that will show you a nice webview to see where you locate it, which mirror are you using currently and a bunch of stats that could help you diagnose the goal? Is you should be able to be autonomous to determine if there is an issue on the jenkins infrastructure or on your network or on the mirror and switch mirror? If it's the case next slide, please another element is more transparency.

F

This is this is for every person and for the maintainer as well. So we try as much as possible to give you a state of the infrastructure on status.jenkins.io, which is a new web service. The goal of that service is to let you let you know. If there is something going wrong on our side on on your site, everyone is able to contribute. If you detect something faster than us. That can happen and it happens. You can open a pull request to start an incident.

F

You have access to metrics, you have access to maintenance window on the future and each time there is an incident an outage or something we try to be as much transparent as possible and publish this postmortem afterwards.

F

So don't hesitate if you have any issue, propose proposal for improvement because it's brand new we are far from perfect, but we try to give as much information as possible, so you will be again autonomous to determine if it's an issue on outside or on your side next site. Please.

F

uh So the next step for us in the upcoming months will we started and we are going to keep the priority on controlling the costs. So that means diversifying cloud providers to be sure that we we don't have a big outage. We need the blast radius when something goes down and also because different clouds provide different services for different costs. So if we can use a lot, we will have a more sustainable infrastructure in the future and an ability to switch the cost and control them a bit more as much as possible.

F

Next slide. Please also big effort is currently being done and will continue to be done during the summer on ci jenkins io, because that one is responsible for a part of the cost increase. So there are more activity, but we can still control more elements in order to do that, we need to improve measurements, to be sure that we control everything, and so we have fundational work to do.

F

The main idea is to make that jenkins instance, the let's say the referendum for everyone using configures code, automation, multicloud providers next slide. Please.

F

So just one slide to thanks all the contributors to the infrastructure, because we have people working day to day, but we have people doing valuable work. Punctually, weekly monthly depends. So, if you're interested, if you like, this, don't hesitate, we try to be as open as possible. This is a shared infrastructure.

F

Everything is open, except the credential I hope, but for the rest yeah. Thank you very much for everyone who contributed, even if it was one line of documentation, it's still a contribution. If it's reporting an issue, it's still a contribution, we need, you don't hesitate, go visit us next slide. Please mainly you can learn on the static page on your own page.

F

You can check the roadmap. You can join us on irc and you are welcome to join our weekly meetings, so you can find on the link on the slides. The slides have been shared on on the chats. If you want to join us, participate or just spectate, you are more than welcome thanks. That will be all for me.

D

Thank you and yeah. Then we will have some updates from special interest groups uh and details, and the first group.

A

We got invited this ux group so.

D

Team would you like to share what's going on there.

G

G

um So, a few months ago um we got new, build status, icons and weather icons. um That's in the latest, lts 2.289 um uli's been doing some work on bootstrap, five and e-charts five. um I think those are all available now um and some of some of the major plugins are updated to them. Junit is work in progress, but should be done soon.

G

I think um trend charts are configurable for the plugins that have been updated to use the new versions, um so you've got a few more options: kind of to replace the different size options you used to have in the old graphs. um There's a pull request, dashboard plug-in around, which is getting some integrations with different plugins. So you can customize what your pull request.

G

Dashboard looks like integrated with like code coverage and warnings, ng and probably others, there's a pipeline view plugin that I've been working on, which kind of pulls the blue ocean graph into the classic ui um and there's a work in progress, poor crest for um logs as well.

G

um So it's quite nice um and hopefully means that we gotta get rid of blue ocean for some of the people who don't enjoy it so much and recently, garen mogan's done some work on the plug-in site and searching release notes and we had a contribution to give suspended plugin pages. Previously they were just four or four pages, but now um there's a link explaining it.

G

So all some good improvements there.

A

Thank you and for some items uh on this list. If you want to know more, um currently, we have a section on the community in scion which is called showing off, and this is a session where we I can find all the recent updates. So, for example, let's take a look, sean cove and here tim uh talks about lotion, but in classic ui. Basically it is announced- and here you can find some examples and links to the repository.

A

So this is one of the ways for you to get new information, because we encourage contributors to actually post quick updates here about what they're doing, instead of spending much time writing a blog post, etc. Here you can put an update just in a few minutes and we will be happy to use this update to promote the work you're doing in the drinks project.

A

So thanks team for this work and thanks to all your uxc for all the improvements in the recent versions, because yeah jenkins has changed a lot over the past year uh as an experiment. I launched the version from february last year and from today's version one week ago, and they were completely different in terms of how they look so nice work and yeah looking forward to see plugin monitoring and looking forward to seeing the oceans.

A

Thanks tim moving on or.

D

Is there anyone else from the uxc who wants to share the experiences.

A

Sorry, okay, then platform seek mark. Would you like to share that right.

C

Yes, so the platform sig continues to meet and discuss platform-related topics like docker and java versions and operating systems, we've upgraded the java versions and the operating systems inside our docker images. We're really pleased with the results we're getting from the plug-in installation manager tool. It's now included in the docker images for the controller and the documentation. Pr has been submitted by sudakar a new contributor, so we're delighted to have new contributors who are submitting relevant and significant documentation improvements, mac os support improvements have arrived in core freebsd support.

C

Improvements are coming and we're seeing lots of people doing testing and wider use with arm platforms. So it's not just an intel world anymore. Next slide.

C

We'll be discussing in the platform in the summit today upcoming changes that have been proposed relative to java 11.. There are some different choices we need to make in terms of what the path should be forward for java 11. and likewise in september. We expect that the open, jdk project will release their next long-term support release.

C

Jdk 17, jdk 17 will be discussed in that same contributor track, and we have ongoing need for further docker updates, including upgrades to java and operating system upgrades as new operating systems are released and new java versions are released. That's all that I've got oleg.

A

Thank you and the next cloud.

D

Native seek so kara party half and should we will uh give a quick update.

H

Hi, I'm kara de la mark. I lead the cloud native sig and really it's a space for individuals to to be able to bring their ideas on how to make jenkins itself more cloud native, and I don't know why I'm not showing up, but um really you could bring the ideas that you want to work on and we've had a fantastic year so far, because the number of people have brought great initiatives to move forward and we have three of those individual contributors.

H

I'm going to speak to you today we have bartek antonia, bobad and shruti. Chateau verdi and bartik will be speaking to you about the kubernetes operator for jenkins, which he has been involved in leading and also leading a gsoc project. On that kubernetes operator.

H

Fiba will speak about the section client plugin, which he created the poc for and has been working on and um very exciting developments there, and then he will also speak about the cloud events plug-in, which uh is now a gsoft project. That triti is our mentee for and all three of these individuals will be leading breakout sessions later in the day. So please do join them because they are bound to be incredibly interesting. The cloud native sig itself meets bi-weekly on fridays at 11 um utc.

H

So please do join us very open meeting we're very friendly. You can just bring your ideas and chat with us. Ask questions get involved and we hope to see more of you. Thank you.

B

Thank you so much.

I

Oled um hi everyone, so I would like to keep it simple uh now because we have another session later this this afternoon. So we can. We can deep dive into particular topic and engage with the with the wider community, and um I would like to focus on three key areas uh which is roadmap and future releases.

I

So what team is currently working on um then move on to a summary of the feedback, mostly from our community talks and some dedicated uh feedback form to understand how people use jenkins in a cloud-native environment, mostly in terms of kubernetes and and how we can improve the overall experience and finally talk briefly about our involvement in the community space, because this is the very first time we are part of the contributor summit and recently jenkins operator itself became official subproject of jenkins. So there is more yet to do the more yet to come.

I

Could you please move on to the next slide?

I

Okay, so one of the the major uh things we are currently working on is a new api schema for the junkies operator. We want to basically provide more granular custom resources instead of having one big configuration file. We would like to have more smaller individual parts which can be configured independently, uh just to increase overall developer experience, as well as um think about modular code base, which is easier to extend. We can implement uh support for other platforms.

I

More capabilities people basically can contribute in a better way, because we don't have this monolithic architecture and it's fully modular, and apart from that, it also helps with testing strategy, because we can test individual pieces running in individual reconciliation controllers when it comes to operator pattern and beyond that, we also want to uh provide better integration with third-party tooling, especially in area of cloud-native technology landscape.

I

For example, cycle injection allows us to integrate jenkins with service mesh technology, for example. Right also, we want to stay on top of the technology, upgrade sdk all components and and bring back fixes and improvements over time. um Could you move on to the next slide?

I

um Speaking of uh summary, um this is the slide we first we wanted to understand how people use jenkins operator and, as you can see, uh the junkies operator is mainly used in big enterprises.

I

They have quite special problems, mostly related to multi-tenancy, doing everything at scale and also some security and compliance uh issues. uh We want to simplify when using netis operator next slide, please um what are the main problems opera operator already solves. um So this is the plugin management and dependency health problems, so we can install plugins programmatically. We can manage versions thanks to this plugin installation tool, which is part of the implementation.

I

Also, it brings a more scalable and better solution for developers. So the entry point uh is slower for developers. They can easily build and deploy fully operational jenkins at day one.

I

Moreover, they can configure everything in code which opens room for github's model, uh immutable infrastructure and basically scaling devops team in general next slide. Please, and of course, we still have plenty of things to improve and I would appreciate to understand what are you? uh What are the pain points uh based on the summary uh from from the feedback form? Backup advisor is still the thing, especially on demand how people can control backup and restore how backup resistor strategy should work in kubernetes um in the jenkins, more granularity for a configuration asset.

I

So how devops team can configure individual pieces using configuration as called and again plugins management related to versioning compatibility? Caching? All of that we still need to work on that and I believe um there will be a the last slide. uh Could you move on like yeah? So um last but not least, our involvement in community space, so, as mentioned earlier, yankees operator became official sub project.

I

So we are fully committed to be part of the of the of the governance board, of the junkies to jenkins itself be active and also we take part in google summer of code.

I

We try to implement additional security capabilities related to installing plugins, basically would like to have um an additional security layer when installing plugins and monitoring security vulnerabilities on the fly and- and also it's worth mentioning um that we plan to have closer cooperation with continuous delivery foundation, um I don't want to disclose any any particular information.

I

um Just will keep you informed as soon as we finish the formal steps and, of course, a dedicated category on this course uh feel free to bring some new topics, especially let us know how we can engage. Maybe there is particular topic. We can talk about community uh meeting, so that's pretty much all. Thank you very much.

B

Ask a quick question.

A

uh Maybe we could postpone it later because we are already running out of time or doing them asynchronously. If you don't mind.

D

J

Okay, so hi guys, I'm vibhav um so before we move on to like the plugins that the cloud native site is actively working on right now, I would like to kind of talk about this uh idea of factors of interoperability, based on which we have chosen to work on these things.

J

So the first thing I would like to talk about is like factors of interoperability means that you know you. When ci cds need to work together, they need to figure out some stuffs like they need to figure out through what factors they can work together. So we figured out that there are two different factors in which this could happen.

J

You could work directly with other cicd tools by in a client based approach, or you could indirectly work through uh events or like data streams which would kind of trigger jobs based on you know, uh just uh events that might happen from other ci cd and you would listen on them and based on them. You would uh trigger your jobs or like pipelines.

J

So this is the approach we went with so when thinking about like what projects to pick up so looking at the examples uh next slide, yeah. Looking at the examples uh we have so for the direct interoperability, we uh we want to focus on tecton because tecton is is now like the first class kubernetes uh ci cd pipeline system and it it provides a really good way to create pipelines which are based on crds and kubernetes, and it is very, very modular in this way.

J

So uh for that we started with the tecton client plugin for jenkins- and this is this- is the first take at you- know: interoperability between uh jenkins and tecton, but from the jenkins point of view, if uh jenkins had to do something similar well, it had to call a tech on custom resource or create a tecton. uh Sorry watching, however, techton had to create uh jenkins job or, like uh technolon, had to trigger a jenkins job.

J

For this uh there is work going on, in which uh the jenkins uh techton should be able to create a custom task which triggers a jenkins job. So for this a custom task controller is being worked on and you can see the link that is given in the slide over here which, where you can get like more information on how that would work, and this is direct and drop interoperability. We are talking about in this case.

J

But what about cases where you know ci cd systems are like so disparate that you know you possibly cannot just uh uh call like in in some different kind of where you possibly cannot just you know, do like client calls, and you would rather uh use like direct data streams or like events to make that happen.

J

So the first go at that was uh using cloud events, so cloud events are something that was already adopted in tecton directly, uh but now there is work going on in the in tecton for creating a new cloud events, controller which will use the continuous delivery foundation's uh cdf vocabulary, uh like the certain vocabulary, which could be used across ci cd systems to kind of standardize.

J

What kind of events happen across the icd systems and you can check out sig events and see cdf for more information, but using this standard we can basically kind of you know, say that okay upon upon this event in jenkins, I want this thing to happen in tecton for upon this event in tecton.

J

I want this thing to happen in jenkins, so for that uh we uh started the gsoc project for cloud events plugin based on the cloud events uh controller that was existing in tecton at the time and slowly we are moving towards you know getting a feel of what interoperability looks like. um There is also a poc which is being worked on in the sig events uh in cdf, so uh which kind of which shows the improbability between tecton and captain, but I hope in time we'll be able to do that with jenkins and technon soon.

J

So I'll talk a little bit more about the tecton client plug-in itself, so it's uh yeah. So it's so the tecton client plug-in here so based on. What's given on the repo itself, the tecton plant plug-in uh kind of allows us to call tecton directly and techton is basically uh based on the repo tecton, provides the ability to create a k-test style resources for declaring ci cd style pipelines. So you can use kubernetes crds to create pipelines and kubernetes, and this is very kubernetes based and what so.

J

We recently have released version one for uh techton and uh we and recently james, stark and also posted a blog post for using uh jenkins with tech, town and jenkins x. So.

C

Like slides, are sorry to disrupt, slides are shifting around all like.

D

Sorry uh am I still connected.

B

D

B

C

B

We're back to the outrage, advocacy. Sorry excuse.

C

B

D

B

D

Yeah I was trying to find the chat because we're going over time for.

A

For something like 10 to 15 minutes um and yeah, I messed up the slides a bit yeah. Let's try to accelerate.

J

Okay, that's, okay! Okay, so let me continue um so we recently released v1 for tecton client plugin. You can go check it out and start playing around with it, but uh there's some interesting stuff. We are working on next for the plug-in itself, so we are working on some uh tecton cloud configurations with which you could use tecton from multiple kubernetes instances and deploy pipelines from multiple on multiple kubernetes instances from jenkins.

J

So that's something interesting we're working on currently and we plan on working on a pipeline dsl uh with which uh writing technon pipelines would should be easier in jenkins as well as, uh and it should be a little more familiar to jenkins users as well to like techno in that way and not have to deal with a lot of yaml and uh the other stuff we are working on is support files with multiple types that that means currently, when you uh give a file in uh creation for in the plug-in, uh the file cannot contain multiple types.

J

It only has one. It only can have like one resource and not multiple resources and then currently we have catalog integration using jenkins x, uh which is really cool. uh Techno catalog is basically a catalog of different tasks and different tasks, which you can reuse and you won't have to write your own task. So it's basically tasks which have been vetted and created by the tecton community, and you know they keep improving it from time to time and keep releasing new new version of the catalog.

J

So it's a very reliable source to kind of start using tasks. uh So we would like to have better catalog integration so, like uh the users wouldn't have to figure out where to get the task from like they should just have it in their hands and jenkins. So and the last thing is cloud events configuration.

J

So this uh would be quite in the future, but as the cloud events plugin grows uh and we are able to figure out, you know how to create things and everything, and we should just be able to map uh eventlessness from uh techton directly to jenkins uh sinks, and you know kind of figure that out easily without having to do a lot of back and forth. So yeah, that's about it for techno. Thank you for listening I'll now, pass the torch to.

D

D

Okay, so let's try to accelerate a bit shortly.

D

D

Okay, this is just that we move forward and then we can return back to that or we will have a breakout session if.

D

C

Yes, please so, very briefly, thanks very much to those who've been involved in improving documentation by adding embedded search thanks to the she-code africa contributon for their work on pipeline help and syntax and snippet generator, and thanks to diraj for becoming the new maintainer of the weekly change log. Next.

F

C

uh We're grateful for plug-in documentation as code, specifically the contributions from gavin and from oleg and for the embedded search next slide and really want to thank those who are involved in chico africa, including zenob, and the five contributors from africa, onigna sharon, esther cynthia and lucy, and our mentors kristen and meg and angelique, and oleg thanks very, very much we're so grateful for what you did next slide. Thank you and we'll we'll be doing more of that, and that's really it for me. Oleg thanks a bunch.

D

Okay, thank you. Yeah I'll also do a very accelerated update.

A

For books and a fish uh so yeah, as I said, we do a lot of social media management, including twitter, linkedin, youtube reddit. Please subscribe to any channel, we do a lot of automation and we are looking for contributors. So here I, like writing great content. Please do so and yeah. I also have this course. So this course for us is not just the discussions. It's also a lot of special opportunities for seek like doing announcements, doing easy, blog posts, uh showcasing achievements like represented today, introducing contributors or think contributors.

A

I will do some technical talks, but yeah. We intend to use the discourse a lot more. We also plan to keep uh working on teach programs, so we had cdcon this week. There will be devops vault in september. Maybe there will be other major conferences we participate uh still yet to be seen. We also participate in outreach programs, so she got africa has just finished and we have ongoing a google summer of code. uh Oktoberfest is in october as always, and we participate and we plan to run one or two lfx mentorship programs.

A

uh You know autumn uh basically, self-funded to address topics which we cannot address uh through existing platforms and uh also thanks a lot to educational issues. So uh lisa will you have been working on this topic and also vadik who contributed to security, uh investigation and analysis with xyl uh students. It's also much appreciated and we invite people to contribute to this area as well. So just uh shout out to african contributors. Basically mark has already said a few words um thanks a lot, and I want to highlight that it's not just about code.

A

We also invest a lot in documentation. For example, xenapp was meant to be in google season of dogs last year. Unfortunately, we haven't been accepted this year, but we will apply again and yeah. For us, it was still a great experience and it provided a lot of opportunities with ship coat africa uh so that we could collaborate between our communities.

A

The german knowledge cleanup uh there is some armature updates, so we defined uh sub-terms for jenkins controllers. So we unlocked the most of contributions. There are more than 50 pull requests submitted over three months and there are more upcoming from what I see in the backtracker and in the conversations, and there is a jab by angelic chart about the inclusion for terminology guidelines and continuous updates so that we actually may have formal standards within our community and these many other projects we are looking for contributors.

A

There is a link here which is pointing again to this course, which provides all the information how you contribute and what types of contributions we are currently looking for, and thanks a lot to everyone who wants to contribute uh for google summer. Of course, this year we have five uh projects uh again, a central framework. We do a lot of office hours, we will do project updates. Hopefully we'll have some talks about gsoc this year and uh thanks a lot to five students who participate so shruti, aditya, hashida, akihara and uh phuket.

A

They work on different areas in the jinx community on important topics, uh mostly called native ecosystem, but also infrastructure, foundation, plugins and very much appreciated, so ishwari back online.

C

Shruti and I were just discussing oleg and and she was agreeable to hold the her discussion on cloud events inside the breakout session- that's going okay! Well, let's.

A

uh Put it to break out and yeah what I wanted to say again that it's a great time to contribute. So there is a lot of opportunities for contributing. If you're a newcomer, please join mark's session from newcomer contributor. If you're experienced contributors, we will start the end user panel, uh maybe with a five minutes break and then, at the end of the hour we will start breakout sessions.

A

I intend to start them on time, so you'll have plugin enterprise, holistic content, ability observability on telemetry, and then we will have more sessions continuing and thanks a lot to everyone who joined the kickoff. We apologize for issues with the webinar link and we will definitely run a retrospective in order to make our events better and also you have a lot of opportunities to share your feedback. I've already posted a link to the feedback form. uh Please use that also.

A

There is this course discussion where you can find, or just use our slack and put your comments there and that's it. So thanks a lot, uh so we can start, but actually we are going to have a five minutes break before the next session. So we start at 20 minutes or at 50 if you're in india and yeah thanks a lot to everyone.

A

Let's keep working together and I will share the presentation, the feedback for me, because many of you may have already have feedback to share.

C

Thank you. Oleg and I've had to revise the zoom link for the newcomer session, so in the five minute break. Those of you who would like to join the newcomer session. If you could just send me a chat message, I'll post the link to that newcomer session in the chat messages as well uh that way. You've it's accessible to you.

A

Yes and uh for breakout sessions, we will continue here. So thank us.

C

Yeah so so, let's see you in five minutes, see you in five minutes thanks. Everyone.

A

Yeah I'll stop the recording because I'll uh start a new recording once we're.

A