 Hey, Nikolai. Hey, Andre. Is it around these days? I think I don't think I saw seeing him. I talked with him. Ah, okay. Yesterday. It was like so. And he says it's pretty crazy situation and he has at the moment. Yeah. Seven impossible to buy toilet paper. I don't know why. It seems like, like, this is a common problem all over the world. I don't know why. Why the toilet paper specifically, but it's really, really strange. I think that there should be some psych psychologist. You know, just trying to. To work on this and try to figure out what actually happens in the human brain in times of crisis. So that they, they go and buy. Yeah. I know that, for example, the bees, when there is fire in the, when they may smell fire, they go and, you know, suck all the honey, honey that's in the, in the, in the hive and try because they, they want to ensure their survival. So when, when they migrate out of the hive and they just want to move the, the honey. But I mean, that's something that, that makes sense. Toilet paper. Wow. Strange. Good. I wonder if people are still joining the other, the other chat at this point in this time or. Sure. We will check. I'm going in your area. I heard from some guys in France that it's mostly restricted to go out without real need. Yeah. Here is not that form yet, but I guess we are, we're going, we're going there with, with full speed. I mean, people are just, you know, they, they just need some restrictions to be put more firmly. Otherwise, you know, I'm just trying to skip each and every rule that is possible there. So once you tell them, you know, you're going to jail or you, you have, you know, 500 euro fine, then suddenly they start thinking more magic. Hi, Frederick. How are you? Hi, Frederick. Hello. How are you? Yeah. How are you? Not many of us today actually. You didn't know Ed will join or not? No, it's not going to be able to join. Yeah. He's not going to, he's not going to be able to understand. Okay. Let's wait a few more minutes. Probably someone also will join us. Sounds good. I think we could start. Does anyone have something in agenda? As I know, VP agent was released 3.1. So I think we could switch to it. Probably. Is that correct? Is that currently in progress or are we currently switching? I don't know when to switch it to 3.0.1 and I think we need, I just checked it a few minutes ago. We wanted to release and we release it. There is one interesting features. Probably we need for a new SDK from it. It's ability to pull the metrics using a request without a real pulling, but just send request and receive the metrics. I'm not sure about the stability of VP agent. Let's release. So probably we can try update and monitor it and check how this will be going. Yeah, I think that's a good idea. Okay. I will create a issue in this case. After the merge, I will be able to finish the metrics pull requests. It was ongoing and waited for this function to be released with VP agent. I think you have some progress with a wire guard. And we found that modification of a pure modification of a packet. VPP component is not fully logical. And at the moment, we think about having a plug-in for a VPP to perform with required functionality. Could you please give us more details about it? Oh, yes. On last week, I tried to somehow patch I've packet to make it work with where free, but I did not find any easy way. Actually, we can patch it, but it looks like that a more easy way is just to add a plugin for VPP. I also have some opinion and I have some options and other options how we can fix this problem with incompatibility of wire guard and VPP. And Frederick, do you mind to discuss it by Slack? Yeah, I'm happy to discuss it over Slack. Oh, fine. I'll provide details by Slack. Also, last week, I worked with DNS simplification. We have discussed about these resets. I have provided PR, but I also went to cover one conflict case. Actually, I found one interesting case which is not covered yet with this simplification. And I asked it about my solution for this case. And actually, it's not soft. It's my suggestion. And I will complete my PR when I look at my suggestion. Mostly, that's all from my site. Okay, nice. Yeah, well, send me the details on what you have. And I'll take a look at it. Yeah. Okay. Yes, sure. Also, on our site, we started to work on its MD and S manager. And we tried in a few days to fill the number of issues with tasks required for different chain elements to be put into different SDK parts for business manager to be alive. The most interesting, I think it's a forward there, choose a chain element. This is a part of a new concept from it, about replacing a forward there with just a point with a functionality of a forward there. So probably right after it, it'll be more easy to construct the final NS manager application on a new SDK approach. Yeah, that makes sense. So I'm being at our jar going to try to work in order to get some more clarity on some of that stuff as well. Because I know that there's still some, some gaps in the concepts. Yeah, that will be nice. Yeah. So more likely it'll probably be later in the week rather than sooner on that just as a heads up, but we'll work on getting that information out. Sounds good. Do I have anything from your side? So let's see, in terms of development, I've personally have actually been sent home from my job because of the virus that's going around here. So I would have more time to spend in order to help you all out as well. So I should be a little bit more proactive on some of the, on some of the, the code base. With that, I think the main, the main thing that we need to do is just get you all unblocked on some of the, some of the gaps and get you all unblocked with, with things really towards, towards wire guard. And so once, so once that is, is back, then I will, so he's not going to be in today, but he should be back. I think either tomorrow or the day afterwards. And once he's, once he's back, then I will, I will work with him in order to try to get all this, this information out there in a, in a consumable format. Yeah, sounds good. Also at a moment previously we discussed about switching from current monetary approach of multiple GPC servers per every client and then point to a single GPC server. And probably I found the solution and working on it. But from the implementation point of view, I'm not very happy how it going. My idea is to use the bidirectional streams of visual PC to end point after caller, register in SE, do call a special method to handle requests and send responses back to an S manager. In this case, we could use just one server socket. And we will need for the endpoint client sockets. And if we'll need them, maybe if we could still have a device plugin provided by a forwarder, but for a generic NS manager, we will not need the device plugin at all. I think it should speed up the startup of the clients of endpoints and so on because the device plugin is pretty slow actually at the moment. Yeah, that makes sense. So you said the implementation, like you've had a first pass on it, but you're still not happy with the words that? Yeah, yeah, yeah. So initial idea was to like client connect to the server, identify this client and create the pure GPC connection and able to call back any of the GPC APIs. But I actually could not find a very easy way to do so. So it's probably with modification of portable generators, but I think it will be too much for such. Yeah, this is a problem with the GPC itself where GPC is designed as a client server relationship, not a peer to peer relationship. And so you may run into problems with that. That doesn't mean that there's no way to make it work. There may be a way to do so, but I can't think of an easy way to do so off the top of my head. Yeah, the bidirectional streams are actually allowed to do so. So it's some complications on the client side when I need to handle all of these calls. So generic calls will work without problems. It could be problems with streams from a server to client, but I don't think we need such for when point cases. So probably it will be okay just for our solution. I will prepare a spec after we'll be mostly okay with my experiments. Yeah, cool. Thanks for looking at this as well. We'll speed up. Yeah, speed up and simplify. Most interestingly, it will simplify an SM server since we will not need the workspace concept. In general, you can as manager. It will be just one server socket. And I think it will be much easier to handle and test. Cool. Yeah, definitely get us the spec earlier as well. So we can, we can get a census to use to what it's doing. I will, we'll try to, we'll try to help you identify the, the corner cases on it. Yeah, okay. That was good. Nikolai, do you have anything from your side? No, not really. Yeah, at the moment we have tasks we need to work on. So if nothing more important to discuss, we can just go out for sometimes and join on a community meeting in the 15 minutes. Yeah, I think that's fine. Yeah, see you in the 15 minutes case. Thank you all. Thank you all for your play. See you. Bye.