 Good morning. Hi, Ed. Hey, apologies for the confusion. Good morning. Hi, Ed. Hi, guys. All right. Yeah, apologies for the confusion. The new security features we have don't start the meeting until the host joins. To join as the host right now, I have to log into the network service mission account. And for reasons that I do not understand, my personal account was allowed to join. So I was sitting there alone in the room, wondering where everyone was. So we still need to sort out some of the details of all of this to smooth it out. Most particularly, how to let people join if they just have the host key. Because that would then allow us to spread the joining out a little bit more and make it easier on everyone. And then I wouldn't actually have to, let me rename myself here, log into this account to come in. So good morning. How are you guys? Yep. Yep, I'm fine. How are you? I am feeling triumphant this morning. I just finally pushed out the to-go stuff that I dropped your note about, that I'm super happy about. Awesome. So let's see. Shall we get going then? Let me see if I can bring up the, so I guess the question is, so like what things do you guys need at this point? So I know Denise, you need to put a couple of questions asked answered by me from BL3. And I've noted those. And now that I've fixed my head up to go, I think I'll go, probably get to those fairly shortly. Anything else that you guys need? Probably to look at few progress into SDK. From my side, mostly. And that's all. Okay. Apologies, I've fallen a bit behind. Let me take a look and see. Yeah. There are definitely some that are falling behind. I'll go catch up on that this morning as well. Cool. I will point out, go have a look. I know I sent most of you this morning the to-go stuff. Go look at to-go. It will allow us to have extremely clean and developer friendly Docker files. So basically we can have Docker files that look like we're just doing normal ghost stuff. But number one, the caches will always be warm. And number two, if we have local replaced directives, those will also work inside the Docker container that we built. Which makes development tractable. Nice. You look. Yeah. Ended up being a bone simple approach. Awesome. So anything else this morning? Otherwise I'm inclined to yield back the time and we'll meet up with the community meeting at the top of the R. Mostly I'm okay. Just you, SDK for requests. I would like if somebody will look. One is to slightly modify aspire stuff to allow temporary folder. And registration to temporary folder. Also it allowed to use it. For example on Mac OS if aspire agent and aspire servers compiled. Okay. So it's possible to use locally all the tests instead of running into the Docker. Yep. And few more. You know, it's the kind of stuff. At the moment I'm working on integration and call back stuff into SDK. So probably we'll create a pull request tomorrow and integrate all of this stuff into send the manager application repo. Excellent. Cool. All right. Anything else? Just few probably questions. Interesting. Related to registration of the end points. At the moment I have doing allow to specify few a listing on your eyes to send them NS manager. And for any TCP ones plan to do a register of NS manager into a registry. And interesting question if we pass multiple TCP address to listing on. For example, we could specify few different IP address to listing on. Potentially it could be interesting. And do we need or do few times do register into registry with different IPs. I think we do. And interesting question for end points in this case because end points bind to one NS manager by its public IP, public IP address at some moment in the registry and what to do in this case because it's still one end point but potentially it's as above as two public addresses. Okay. I see your point there. So effectively a choice has to be made about which address to register on with the registry. And so my baseline point is the address you used to register with the registry may actually just effectively the way we traditionally been doing it is that you run your because you're running on a pod normally you normally have just the one pod IP address available to you. And so it's not really but an issue. You know, effectively we don't even we don't advertise based on where we're listening we advertise the pod IP key. Right? Because that's where, because it's not where you're listening to matters it's where you can be reached. Yeah. For a moment I could just limit for example to be one or two connections and if user pass two, it should be Unix and the TCP for example. Yep. And the TCP on will be used it for a public and more interesting question here at the moment in Manarepa for a public IP we do not register registry GPC service and some other services. So it's possible to register in point only with Unix socket. Do we plan to have a similar approach a new version I think? And if you- Yeah, I think essentially it comes under a question of policy, which is effectively if you're a network service manager as a matter of policy, you're only accepting incoming registration requests from a Unix socket as a matter of policy. Yeah, okay. Sounds good. It's definitely an interesting question. And the thing is by moving into policy it makes the code more generically applicable outside of the community space where you may have different kinds of behavior going on. Yeah. If policy it sounds good to me. All right. Yeah, cool. Okay, in this case. Anything else? All right, we'll see you guys later. Yes, I'm sorry. Please, please go ahead. I can't provide my status if you don't mind. Oh, that's marvelous. Please do. Oh, okay. Currently I'm working in VAL3 NSEe direction. In detail, in first I have provided VAL3 staff requirements diagram. Oh, yes. Let me show the screen. Yep. And also I can put a link into the chat. If you have any questions or thoughts you are welcome. Okay, let's see. Are you seeing the screen? Yeah, it looks like you're talking about, is this a diagram of? Yes, it's a diagram of VAL3 staff requirements. And currently I'm working on a VAL3 I working on memory registry elements because of we need to somehow test, for example, on some manager. We didn't remember register chain element for registry is completely brilliant because yes, you can use that for very localized testing. Cool. Yes. And I will have provide power tomorrow for this chain elements. Also I have prepared diagram for VAL3 interaction, diagram of interaction. Let me show it. I have provided also the issue into SDK repository. And currently this issue is blocked. You can see this by this diagram. Yep. And here I have provided simple diagram of interaction. It is not final variant. You can take a look and provide your source or source. This is excellent. I apologize. I've been hyper focused on code in the last week. I will definitely go get through this. And I would encourage other people to go take a look at it as well. Cause this is excellent work to sort of sketch out where we're going with the VAL3 staff. Cool. And mostly that's it. I just focus it on, memory chain elements for registry. And that's it. Cool. Now the memory registry stuff should make a bunch of testing that things easier. Awesome. All right. Talk to you guys at the top of the arrow.