Everything EOS EOS Leap Upgrade Group, 3 Aug 2022

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: EOS Mandel Weekly Meeting #26 (Aug 3rd 2022)

Description

This week the Mandel huddle group updates the community with testing on Jungle 4, kylin Faucet tool, 3.1 upgrade guide and more!

00:00:03 Recap and this weeks topics by Daniel
00:01:18 Mandel software update by Stephen
00:04:17 Testing on Jungle 4 by Michael, Daniel
00:07:50 Kylin faucet by Daniel
00:10:43 Power Up update by Daniel
00:11:06 Mandel 3.1 upgrade guide and survey feedback by Daniel
00:18:55 Blocks log archive by Ross, Michael, Matt, Corvin
00:29:39 Stress testing by Michael, Ted, Daniel
00:34:00 Go EOS!
00:34:10 End

🟢 In progress action items are on track:
Mandel software development #41
RC3 release is imminent. It will likely become final release if no issues are discovered after a week of testing on Jungle4.
It's possible RC4 will be required before we go to final
Next release (RC4 or Final) would likely follow on August 3 or August 10
EOS Support has shared responses to the survey requesting feedback on fill-pg:
2979424562146452305fill-pg-feedback_update_20july22.pdf✎ EditSign
Test Mandel on Jungle 4 #44
BLOCKCHAIN_PARAMETERS has been activated
15/15 BPs have signed MSIG to deploy updated system contracts on Jungle4. Pending execution.
Focus for testing on Jungle4 this week:
deploy system contracts
BPs should upgrade to Mandel 3.1 RC3
Other EOSIO chains will be participating in testing soon too
Testnet strategy #40
EOS Nation has created a faucet smart contract and back-end for Kylin:
Newly released 🚀 (Kylin support)
EOSIO Faucet API
https://faucet.eosnation.io/docs
Create account using public key as active & owner permission.
Send tokens to receiver account.
Generate EOSIO private/public key
Feel free to test it and provide feedback if there's any issues.
EOS Nation is preparing to activate PowerUp on Kylin
ENF is working on an open source performance harness and load test framework
Communicate upgrade plan broadly #50
The Mandel 3.1 Upgrade Guide has been published
We've received feedback from some exchanges and BPs that they are still using the depreciated V1 history solution.
Recommendation is to use Trace history: https://developers.eos.io/manuals/eos/v2.1/nodeos/plugins/trace_api_plugin/index
This could be a potential risk to the September 21 activation date
We will add EOSphere snapshots and blocks backup to upgrade guide: https://snapshots.eosphere.io/
Participants (19)
Host: Daniel Keyes | EOS Nation
Brendon Ross
EOSUSA Michael
Stephen Diesel
Virginia Glass
Kevin Heifner
Ross (EOSphere)
cmadh - Corvin | Liquiid.io (Corvin Meyer auf der Heide)
Max_Cho EdenOnEOS Member-South Korea
Matt Witherspoon
Alysha Mims
Matt Witherspoon
Jeff Werner
Zack Gall
Genie Kim - EOSeoul# NEOPLY
Ted Cahall
Andriy Cryptolions (Andriu Bobrovych)
Johnsonb - Object Computing# Inc.
junhank | NodeONE

Everything EOS is the official media channel of the EOS Network Foundation (ENF). The EOS Network Foundation coordinates financial and non-financial support to encourage the growth and development of the EOS Network. We’re harnessing the power of decentralization to chart a coordinated future for the EOS Network as a force for positive global change.

// EOS Network Foundation //
https://www.eosnetework.com
https://twitter.com/EosNFoundation
https://www.linkedin.com/company/eos-network-foundation
https://t.me/EOSNetworkFoundation
https://discord.gg/eos-network
info@eosnetwork.com
#EOS #EOSIO #ENF

A

Welcome everybody it's august 3rd uh today is the 26th time we're meeting to talk about the mandel upgrade uh quick recap: we've got last week we checked in on mandel rc3 was not yet released. At that time we were aiming for towards the end of the weeks. We'll get a update there. Today, uh dune 1.0 rc1 had been released.

A

Looking for some feedback from developers who are using it, um testing is still in progress. The the focus for this past week was to get the cdt testing and get contracts activated on the uh on jungle or the contracts deployed onto jungle.

A

um We've been working on getting kylon in a better testing state, we're working on a on a bot to enable a faucet. Essentially, so we do have more news to share there more progress there um and last week we were looking for feedback on the 3.1 upgrade guide, which has now been published.

A

So yeah lots of progress has been made in the last week. Let's go a little deeper into each of these topics. um We can start with mandel software update. Can we get an update from enf on how that's going.

B

Absolutely so a bit more progress to report there uh so kudos to the team. uh Late last night we got the last pr that we believe needs to be merged in for a final, stable release.

B

So we haven't discussed uh the details of next steps yet today, but what I anticipate transpiring is that uh we bump that version today get some updated release, notes out and plan to proceed moving forward from there. So um all systems look to be a go and uh really excited about that.

A

So that's gonna be rc3 or final.

C

Yeah, that's all. I was gonna ask yeah.

B

The the belief is that it is final um so that that that's what we'd like to move forward with, uh but I can provide an update if there's any concerns uh this took place at like basically seven hours ago. So I I do need to check with the team just to make sure that they feel fully comfortable. But the tone of it sounded like we're final, um so that I mean that's how I'd like to present it today and we can update as needed, wouldn't.

D

We call it rc3 and then after it's been tested for a week or two on jungle, four, then we just rename it final.

B

Yeah, I'm I'm fine with that process. I've not been part of the rc release process going to final just yet, so that's probably the more intelligent way to voice it ted. Thank you.

D

Because you can't be sure it's final until it's final.

A

Right until we make sure there's no bugs in it, okay, so rc3.

A

Release is imminent.

A

uh Will likely become final, assuming no issues found on jungle.

A

Is that captured accurately.

B

I think that's fair to say, yep.

A

All right anything else to add.

B

I think that's it for now. um I we've not seen anything super critical come in in terms of any uh bugs across the other repos, but we will continue to monitor.

A

Great good progress so we're uh we're close, we're close um okay. I guess I'm guessing there's no, nothing new to report on dune 1.0, unless uh I'm not sure if anyone has any feedback to share based if they've been playing with it um anything they want to share here, or maybe we've received any responses to any of the surveys.

A

I don't see randall on here and I didn't get any any news from him offline, so we I'm assuming no new feedback there we'll give a moment in case anyone wants to chime in all right um testing on jungle. Four. So, uh as I said, the goal for this week is to get the contracts updated. There is an nsig just a moment ago is at 14 out of 15, but I I heard michael is in the process of oh. We got it.

E

A

E

A

I don't have uh my jungle wallets set up here, so if someone wants to do the honor of executing it live, live on the air here sure.

E

A

E

Ahead and run through them and execute them, but just be on the lookout. If we break anything.

A

Or maybe not, maybe it's not good to do it now, when we're on a call, maybe.

C

I'll see you wait for button.

F

That that was why I.

E

Left it not executed, but I'm all I'm all for smashing the buttons.

A

Jungle is there for breaking right. That's that's the.

E

A

But uh so good progress, though so it's we're just an execution away from deploying the contracts onto jungle. Obviously, we hadn't had an opportunity to test them yet um so we'll we'll continue to make that a priority for this week. um Is there anything else that we want to be? I'm guessing. You know with rc3, coming up, that's going to be another priority for testing, make sure we get everyone updates to rc3.

A

F

We're going to.

A

Deploy smart contracts or system contracts.

A

E

uh I had relayed to some of the other networks that to try to test the process, although I don't know it may still be a few more weeks before we're there, but I did at least relay out to some of the other networks to try.

F

A

So we'll have uh other flavors of usao.

F

A

Okay um and when it comes to cdt, is there anything in particular, we need to be doing to put that through its paces. That's more of a smart contract developers task, whereas contracts is a bp test. Is that essentially correct.

E

Yeah, for the most part I mean you know, bps may use cdt to verify contracts or whatever, but I mean in my opinion, obviously I'm not the dev, but most of the time they see differences. You know as it's evolved like. I think there was some size difference in in a couple other versions and things like that. But yeah I mean I think it either compiles or it doesn't.

A

F

We'll uh we'll keep.

A

uh Testing right on track, there.

A

um Testnet kailyn, um we made some progress there. I think you may have seen if, if you're, following in the mandel chat or in the kailan chat, um I'm going to update this with the details too. It's not there yet, but denny put together a handy little back end tool that you can anyone can access. You don't need a kailin account to be able to create a kylin account and issue yourself, some tokens, you just call this little back-end tool and away you go.

A

Let me share it on the screen here for those who haven't seen it yet.

C

Okay, with a little run for its uh money, a couple hours back and it worked great.

A

Yeah awesome yeah, I tried to test it out. Myself seems to be working, so you can generate a key if you don't have one already. Just by hitting that button there and you'll get a public and private key.

A

E

Save it because that may be the last chance you get right.

A

Then you can create an appeal.

F

E

For terminal history,.

A

E

A

I won't go. I won't go through the note.

E

A

Put it put in a key put an account, execute it and you're it's going to auto instantly create an account for you, and then you can issue yourself, 100 tokens at a time with uh this call here.

E

One cool piece about it is it's also: it's a smart contract feature action, I guess technical, but you can trigger it via just the smart contract level as well, obviously for ease of use. That's where that front or the back end whatever comes in, but it is actually just a raw smart contract. So if you already have an account- or if you have a process, you could possibly use it.

A

So yeah you will you would you do need a kylin account to interact directly with the smart contract so right? So if you don't have an account, you can get around that catch2 of needing an account to create an account.

A

If you have an account you're comfortable, interacting with the smart contract, you can do it direct with the smart contract. The other great thing is, it's super easy for us to now we can take the smart contract and back end, and that's one thing denny will work on is why don't we use this on other test nets? We can put it on jungle. We can put it on other flavors of eosio, so it's been made fairly. General purpose. All.

E

Right, we can throw it all over the place holler at me. There.

A

You go yeah, you just gotta load up the smart contract with some tokens, so whoever has kind of the access to all the tokens can can load up that account with the tokens. And now you got a faucet.

A

So that's that any questions or anything to add on the faucet next step is going to be getting power up, um upgrade updated on kylie, so we're working through what exactly needs to be done to get power up, updated, looking back through what we did on mainnet, I'm pretty sure that we were the ones who put up that m6. So we're just we're going to just repeat what we did there on kylan um we also made so last week I shared a draft of the upgrade guide.

A

It's now been published on the enf blog and we've already started passing it around. I know that the eos support guys have published the survey for us too, so you can find the survey right in there we got some feedback. We added the the questions that you had suggested ross um virginia. I got. I got feedback that your feedback was very helpful from uh from andrew. So thank you for, for your help. On the edit there sure my pleasure and um so yeah it looks, looks like things are good.

A

uh We did get some feedback as now now that we're spreading the word more and sharing this the timelines sharing the kind of what you got to do we've been hearing there are some exchanges still running the the depreciated v1 history um I mentioned last week. I believe that there's also some bps still running that which could be a risk for the september 21st date.

A

If the next day, the exchanges that are running v1 are going to have to kind of re-architect their processes to find another history solution, um we're suggesting that they use the state history as an alternative, um but so it's so far no nothing put no pushback in particular on the date, but there is some. I guess some complaints that uh oh we were using that v1 history, that's going to be a challenge.

E

A

Was somebody in.

E

The channel this morning, actually that I was talking to and trying v1 history, state history, confusion, layer, one layer, twos, so yeah I mean it's starting to percolate its way out there, but um dude. I don't. I don't know if theos itself has a preferred recommended exchange integration. You know relaying a few. I knew of um I know. There's history, there's state history.

E

It sounds like the recommended path. Is state history not.

A

Let me make sure.

E

I get that right.

A

Let me make sure I got that right. Trey story is the one that we're suggesting that was, that was a recommendation from matthew.

E

Correct and I'm not as familiar with it, but I know it's a little more manual and whatever, but that that may be getting crossed up a state history as well. For obvious reasons, it sounds like both are possible, but there's a couple different ways to do it. Anyway. I know on all the different eosio chains they kind of deal with that same thing of of how to monitor the chain. So that's the same thing.

A

Anyway, based on my understanding, bart built, trace history, specifically for an exchange exchanges years and years back, so it's it's suited for the needs of the exchanges um we're going to make. I'm not sure we're double checking to make sure that that is mentioned here as a resource.

E

Do you have any examples or use cases, and I haven't had a chance to tear through it, but that seemed to be what the person was kind of looking, for which I I mean.

E

That may be the raw documentation but, like example, uh more than just syntax, but anyway I that may be. You know some of the burden we're trying to overcome because they're starting to trickle in which is good. But I don't have a lot of experience with it. So I can answer v2 and hyperion and state history, and all of that, but yeah trace api and um yeah, even chronicle and stuff like that anyway. Something to think about, as we get more of the exchanges coming in.

A

um Yeah, like what kind of uh examples anything I can call to action, I can put out there to create an example. Besides this documentation or.

E

I I I can't, because you know I'm not really the dev. It doesn't make a lot of sense to me anyway, at some points, but that was there was definitely some confusion. He was from an exchange, I don't know which exchange, but he was he was coming in and he was dealing with hey. We use v1 yeah, not anymore, you don't so and then hyperion and state history, and then that matthew actually reminded you know of trace api.

E

So anyway, point is even I at least myself, I'm a little confused on it, but some direct guidance or you know like there's some people, that kind of specialize in some of those things. But I just noticed that at 4 am in the morning and did my best to try to help.

A

E

And for my tinkering with trey's api, I mean it takes a little finagling. You have to kind of hard code what contracts and abis I mean yeah plug and play.

A

Does it does anyone have any experience like yeah? What what would it take for an exchange? I guess it depends on the processes. uh Corbin.

E

Did you guys try to trace.

F

E

I I was just curious. I know you really stress, tested trying to stream data out of the chain. Did y'all ever try the trace api for any of y'all's contrast. No, no.

F

I mean it's it's enabled for for our uh api nodes, but we aren't using it well, the.

E

F

E

Little different but yeah I mean, if you're not using it, you kind of got to configure what contracts and provided apis is abis. So it's not like state history, where you just kind of ingest and digest, but anyway it's it's a little different beast and I think gray mass is probably some of the guru in it. Sorry.

E

um And then a few exchanges, but if that's going to be our recommended path, I don't know maybe some more information about it, because the first one that trickled in he was definitely confused. Yeah.

A

Okay, so I've asked andrew who wrote our upgrade guide to take a second look at the recommendations. We've we've outlined in that article to make sure face trace is is covered as the recommendation and we'll make sure that it. You know that we've made.

E

The recommendation useful, you can only try so much, but you know I know matthew's a big technical writer, so you can. You can only lead a horse to water, but I can at least relay there was definitely an exchange running v1 that was struggling with what to do now.

A

Right right so um do we have, I don't know if it's too soon, to tell how much of a risk this is to the timelines right at this time, maybe just something we keep monitoring.

A

um The survey is out so we're gonna. We should hope to start to get some more responses from people. um Some of the questions are like what what types of apis do you require? What are you currently using, um so that should help inform us more on how many people are in this state and we'll make sure that, as we approach the deadline, maybe we'll follow up with everybody and get their latest status update and then we'll have to.

A

You know, factor that into our decision when we have our go, no go call for the activation on september 21st, um but for now I think we should maintain 21st as our target date.

C

um Yo daniel, you obviously saw my message if you could put our snapshot server up there as well. I don't know if it's possible to edit the article again, but the other one is more of a question to everyone here, I'm putting up a blocks log archive, even when it's compressed it's going to be like over a terabyte. I think it's compressing now. um Do you think it's valuable to have that available to everybody?

C

You know I mean I suppose, we've got our snapshots, but you know: should we be serving a block slug, you know for them to do restoration.

E

From the aspect of just in general as services to the network, yes, I think a lot of people do try to sync it, but it'd be nice. If you could say hey, there's a way.

C

Yeah, I just think it'll take a while right now, I'm just saying for this I mean: has anybody got any? I know we often say that um you know unless you replace something yourself, you never sure, etc, etc.

C

um You know, maybe I mean I I'm gonna host it there and I don't know if we should be encouraging people that had uh done a complete reply from version uh 2.1 that needed to go and do a complete replay to go back onto the mandel track. um You know it'll be available for them. I.

E

Forgot how painful that rollback was because yeah you can't.

C

They're, all blog anymore,.

E

C

Yeah you can use it, you can use it if you upgraded and went back, but if you went from scratch correct.

E

There's a lot of people on 2-1 I mean I I'm floored that you know have I not floored, but there's a lot of people that are going to get caught by that. That's that's true! You either need to resync it or replay it.

C

Does anybody in the group here have a an opinion on using a backed up blocks log?

C

F

A

I think there's value in offering it not everyone else. People have to different people have a different level of risk in their business that they want to. You know they they're willing to trust a trusted bp to offer. You know, use their blocks.

F

A

They're, I think I think there are people who would use it.

F

A

F

Think probably.

A

Bps should probably, ideally you know.

C

F

Their own, but usually not anybody else, yeah.

E

I mean that, to be honest, I mean you can replay the block log in in theory. If you replay the block log, it should go through and validate blah blah blah blah blah. It's the same thing as sync and yeah other than it's not.

G

E

G

E

G

It's not quite.

E

F

G

Yeah, it's not it's not quite the same, unless you add some some additional options to your nodeos command line. Otherwise it does not fully validate the blocks. Oh okay, I just want to clarify that. Would it.

E

G

Like validation,.

E

G

E

It does do light, validation or whatever I'm not even sure,.

G

It does light validation offhand. It definitely doesn't check things like transaction signatures. I don't recall if it even checks, block signatures.

E

One question and I is when it goes to rejoin the chain and start talking to all those p2ps. I assume if it was malformed or you know different, you would be at a different state and ultimately fork out right. I mean yeah, there's a lot that goes into that, but that's kind of the other thought on it is that once you do come back into the network, wouldn't the consensus catch if it were different than the real chain that you're now thinking to maybe not.

G

But I mean I mean uh the potential problem here would be something like uh well there's actually a couple things you could do, but the an easy one to explain would be like if you somehow coaxed. You know a majority of the bps to use a like a malicious log or something like that. Then yeah then.

E

But you're at the same mercy of the p2p nodes that you're trying to sync to validate right. I mean in theory so well.

G

E

Side conversation, but no.

G

I mean because again, if you're, if you're syncing, you're doing full valve validation playing the log with the default options, you're not doing full validation, full validation, but there are command line options. You can use that. Well, I mean effectively not trust a block clock. It will do full validation on it, so it is possible. Matt would.

C

You are you happy with like hearing that there are other bps that are serving up the archive block log for people to join quickly. I mean: does that make you happy or do you think that's a bit of a problem.

G

I'm I'm not prepared to make a conclusive statement on that on on a recorded call.

E

G

He answered it by voicing his opinions.

E

G

Certainly like, if you don't, if you don't sink from genesis, there's certainly a different uh there's, certainly a different level of trust. But you know what everybody's risk level is: is different. Yeah yeah.

E

Okay and if you're writing those swedish debian files, I mean who knows what you're getting anyway right. If you're writing somebody else's code, a lot of people do, you know, go outside of the bounds. I think it's useful to offer that to people as an option. I definitely think it's it's. I didn't even realize the the validation path was different, so nice, you know piece to help provide that as additional information, but if you don't, it's gonna become harder and harder, and you know what it's like on some of these chains.

E

There are no block two three four and fives available out there. So.

F

Yeah I'll put it there, hey and I'll use it I'll use it myself.

E

I'll grab a copy and we're expanding out our stuff, so you saw our snapshot. Services are really kind of expanding out and that's the next step. So I'm obviously.

C

E

Just need to provide the information along with anyway.

F

Sorry for outside that particular yeah.

A

Any anything there on the question that we should include in like if we are going to be adding this as a resource to the article any is there any advice that we should add as extra information like, for example, these client command line commands that you can make sure. Would I.

E

Would say yeah, maybe somewhere in not obviously maybe the mainstream could make it more confusing, but you know if you're looking to validate the transactions and replay or dealing with it, providing that additional information and even the command line to pass it a different validation, I do think, might be helpful as we kind of scope out what they keep asking, then we can kind of that makes sense. You know more of kind of this exchange on our recommendation and help guide.

E

In my opinion,.

F

Sorry do we do we have block lock splitting in in the latest mental release? Now or not I mean, can you can't you then speed up the replaying speed by by replaying in parallel from different blockades,.

E

Oh, I don't know if it'll speed up like the replay, I'm.

F

Not even sure how much replay works if you're able to if you are able to to replay in parallel from different blockades and then later be able to combine the blocks.

E

Can you replay in parallel, if you're, if you're running a full validation, I don't think you could be playing parallel? Could you you have to do well? I.

G

Think the strategy here would be, you would have you would have. uh I mean you'd, have to have a bunch of snapshots that you um that you're going to use at different block heights, and then you have a you, have a block log that covers all of those slices. It doesn't have to be split. It can just be a you know, one block log, that's uh you know rough linked or something like that.

G

If you're doing it on the same box, anyways uh and then and then you would perform that you perform the replay of each of those slices and then you look at like the uh we call it the integrity hash. You look at the integrity hash at the end of one and the beginning, the other and match them up. um I mean you could definitely do that.

G

It's, uh I don't think, there's any like tooling out there to help automate that, for you, I think you'd have to be pretty uh pretty familiar with what.

E

Or just let it grind and do its natural validation. That's I mean right.

E

Anyway, so the whole point of that was yes, maybe some additional pointers and tips and and whatever geared towards the exchanges with respect to the validation ross, is providing a full block log I'll I'll rip on his connection. Maybe we can offer it as well. I need a good copy.

A

Anyway, okay, so I'm I'm I've asked already andrew to draft up those changes, both the addition of these uh additional resources and also just making sure that we've got the trace history. Recommendation covered and coherent um I'll circulate that among the group here once the draft is there and then we can make sure I'm not I'm not technical enough to provide any specific recommendations here in terms of steps it takes. Maybe maybe um one of you can offer some some guidance there either directly as comments in the document or um on this on this issue.

A

Here, if you want to add any suggestions, that would be useful.

E

And thank you matt that that's great information always about the nun.

A

uh Beauty what else that covers the uh the items that we've been tracking? Is there any other items we want to discuss.

E

I have one, but I'm pretty sure it's an off channel, but this may be the people to talk to about it. I I saw, I think, a github uh issue or something pull about stress testing and, of course, we've been kind of doing some testing on different networks, and I think some of the people may be on here around here, but if there's some conversation about some heavier in smashing and testing things, we've learned things we're doing I'd love to chat with. Whoever might be interested in that.

E

If, if that's something that people are getting to talk about, which it kind of looked like anyway, I figured this might be a good place to put that out there. Catch me on telegram. Please.

A

Cool michael has stress, testing insights to share is that is that the.

E

If that seems to be yeah at the enf level or oci or wherever it was, there were some pr's on some of the github that it looked like that might be of interest, and it just happened to be something that we're actively working on smashing so uh happy to be involved with that got hardware to throw at. It got some knowledge that we've learned.

A

E

Of the places sharing.

E

You can throw a lot at eosio now, making it stay and validate and and be a stable experience is a little different, but it's pretty kevin saw. We brought him in on some questions. We've had for some things about limits and whatever you gotta kind of stretch, some of the limits, but it's it's an impressive code page to smash a lot of transactions in.

D

Yeah we we have kevin, leading up an initiative uh but from the enf to build out a open source performance harness to really do load, testing and benchmarking, and all that make it formalized make it open source. So the other chains can use it etc.

D

um But we think that's a really important thing and that's why I kind of connected him up to michael as well and uh saw that people with uh with the common interests might be able to share some interesting ideas.

F

D

Was the main thing.

F

I'm sorry I'm now starting to spin that up and getting started on it, which is maybe what you saw um yeah we're just at the beginning of that but uh yeah. I. I obviously am interested in any of your findings there and I'm following along closely on the on the telos channel as well.

E

Yeah yeah exactly I know, you're definitely involved, but I just wanted to throw it out there. So if it was more than just you so excited to hear you're kind of leaving that and I'll let you play when you want to burn some stuff or throw some hardware at it. I got. I got some pretty servers.

F

You certainly do.

D

And yeah, we may take you up on that appreciate it. We should take you up on that free servers, I'm all about free servers. Oh.

E

You should see I got a 12th generation, i9 just ripping the snot, I'm water cooled. I mean we're overclocking. It.

C

You keep reminding us.

E

I do in russell attest, I mean our problem. Is you get so fast to hardware us as bps actually have to start slowing down, because there's some counter? You know measures to it, but anyway, that's that's our thing so happy to do it but glad to see an initiative and love to be a part of it.

A

Is there anything here to capture and share in the notes in the summary here? Do we do we want to mention this uh open source stress test? I.

E

Like it, I don't know if it's enough to start announcing it, but I mean it's public on the github, we're.

A

Not looking for any any feedback at this time.

D

It's okay to talk about it because zach zach put it in the um in a tweet. The date it was going to be ready is a little uh aggressive, but it's already been it's been announced by the enf that we're going to do a performance, harness and load testing. So it's okay to talk about it um that we're funding that and and uh we're keenly interested in michael's help and uh and kevin uh kevin won't actually be the lead.

D

A different oci person would be the lead, but if kevin, of course, kevin will be involved in oversight of all of it as well. Very exciting uh project, in my mind, because we'll be able to really talk about um benchmarks about you, know eosio and load testing and we'll make sure that we go forward with every lease, instead of sometimes possibly accidentally going backwards, because we didn't have this type of and we'll make all these results. Public.

A

Awesome is, did I catch enough? Is working on open source performance, harness and load test framework.

F

We yeah all those things.

A

All right, it's in the notes.

A

uh Okay, anything else all right. Thank you. Everybody have an awesome day and we'll check in again next week.

F

Go yes, man, let's go yells eels.