 open to the Evergreen open planning meeting for this week. So I just wanted to show that thing. So the last update for years today, basically the one about the security update that got published with the advisory of the last plug-in round. And the other update is about CentOS that got now chosen as a new base image for Evergreen. I don't expect people to care about that anyway, because, well, we prevent, we forbid workload to run on the master. So people shouldn't be relying on that. And anyway, so the goal, the main goal here was basically be able to update JDKs, which is going to be harder and harder to do on Alpine, even the open JDK team is now not anymore providing JDKs on Alpine. So there are a few issues from time to time about the microlipsi, I'm not sure the naming, but mulepsi, that's the equivalent for the lipsi on other platforms. So anyway, the overall goal is here and in the end, it kind of is the reason why the JDK is unable to easily release the nullpine compatible version, because it's relying on the lipsi. So basically, because of upstream, even the project itself is not going to provide a JDK 11 GI alpine build, and maybe later on is going to be the same quite often, or the future is really uncertain. So I prefer to just switch to this to avoid issues before we go further. So that's been done a few days ago already. So if we have a look at the other things, there are not a lot of big things apart from that, a few usual updates, a few things that should be also updating some things that's going to basically remove a lot of log spam, in some cases. That one is more something I cared about from the logs perspective on the back end, because we were receiving a lot of logs, because the fact we customized the way that I store on the disk, I never agreed. Each time we were getting a warning, because we changed the data when actually it was a fresh startup, so it shouldn't be displayed or something in that case. So, and we didn't change it anyway. So, yeah, we changed that thing upstream in the core. So thanks to that person who basically handled that PR upstream, and then I wanted to just show a few things about the outstanding IDs and initiatives that are ongoing, I would say, on the evergreen side. So thinking more and more about providing an evergreen CLI to kind of streamline and make it easier for newcomers to use evergreen, maybe like have some kind of evergreen provision, something, and it would take care of the right parameters, because, well, it's even quite easy to follow the documentation for running a Docker Cloud-based flavor, as we call it for evergreen, because basically it's a Docker call, but for things like AWS, it's quite much more complicated because, well, AWS, you have to configure a few things, you have to have more rights. And the thing also is that you have to figure up from the thing you would need. And with a CLI, we could be probably much more both prescriptive and drive users, maybe by having some kind of maybe interactive mode where we would ask questions, and the user would just paste the right parameters and we would handle those things for users. So I think there's a lot of potential here for making it easy to use evergreen, which is really one of the core goals of evergreen, making it easy for the targeted audience that that's not generally being aimed at pure experts in general for Jenkins and maybe not even in a system as administrators. That's it. To finally, I wanted to present Nipik, I filed this morning after a few discussions we had in the past. So basically we're receiving a lot of logs in the backend, it works very, very well. And as you can see here, we receive quite a few of them and this is the already triaged one. But the thing is in many cases, we actually receive some things like this one, which is now fixed actually. It's been fixed with the thing I was talking about a few minutes ago with that person whose name on GitHub was Charonbeer. It's supposed to be fixed, that thing. And for instance, that thing that didn't detected the typo has been fixed like a long time ago, like three months old indeed, because I was fixed a few days after it was introduced. But you see that we keep receiving, even four hours ago, logs with that thing. And basically if you dig into an event, we can see that in that case, it got received from an update level that's 173, which is not that concerning because well, not that old, it's a bit old. It's a bit surprising, but not that old because it's only that one. So it's not being updated to that one. But we have many other cases and one in my mind that that's even in 100, not even 100, 66, you update the update level 66. So yeah, we have a few instances that we don't know why are not being updated. So either we screwed up something on the development side or until we are able now to update those instances correctly, that might be a reason. Or we know of definitely some users that have been playing and trying to customize Evergreen where when it's actually not really designed for that. I mean, manually on your side, it's more going to be something like defining a profile or what we call play flavor on the protect side and define something that would actually march or need. And if you try to actually like, we had users ask for it as for to put manually plugins on the disk. Probably it's going to go wrong because Evergreen is going to then each time on each update level is going to remove everything that's not declared in the manifest because we do that, for instance, we'll be able to, when we go from a version and to N plus one and then N plus two, imagine that in the N plus one, we realize that we introduce a new plugin that's actually crushing instances out there or creating a lot of issues. Then we will roll it back by just removing it in N plus two, the plugin that got introduced in N plus one. But in, so that's why we are able to remove things in not only add things. So if people add plugins manually, then we are going to remove it the next time it gets an update, an update. But there, if you hack on the Evergreen and on Jenkins general, you're going to be able probably to find a way to install an additional plugins there. That's absolutely a design, not designed for this, but you could probably manage to do those things or handle the backend some way or something. For instance, but that update level C3.6 is not even a very old instance that can provision like one year ago or something or the beginning of the year at the very early days or something could get broken. It was provisioned around like three months old in November or something. And so yeah, the overall thing is that we can't really do anything about those but we need to start thinking about how we would handle those. So the idea about that is to then first, we want to actually deflect and not propagate those logs into backend anymore but we want to find a way and so to actually reach out to those instances like not really reach out but give away for users to actually reach out to the project and provide us feedback or whatever to try and understand what went wrong because obviously we don't want to design or other way to be able to enter the instances out there ourselves that would sound quite not right. But yeah, so probably the easy way is to in the Evergreen backend expose something like per channel, maybe in the future but for now it's going to be a single version of the last update level and maybe a date or something and say, okay, if you then something in Jenkins that would regularly pull that value and maybe display some kind of wording like what's called an administrative monitor in Jenkins to word the administrator, the users in that instance that, well, something seems fishy. You're running an instance that say has 10, 20, 100 of update levels of being laid behind. And so something is really wrong, please do this and that to and then reach out to us maybe provide some logs or whatever to so that we can understand and maybe first fix your instance and so that it can get back to being updated correctly and maybe understand the core original of that issue so that we can actually fix this so that it never happens again, basically. So yeah, I think it's probably enough for today. If anyone wants to reach out, obviously we're still hanging out on the Evergreen dedicated channel on Gitter and I guess that's it. Thanks everyone and see you later, bye bye.