PeerDAS Breakout #13

Transcript

e e e h

hey guys hello hello thank you for attending the call today let's start with the CL updates um so I think Jimmy can't make it today but he left some message on the on the Discord and say that they identify an issue in the house that about like penalize the peers

sending the colors after they perform the reconst structure so I think that um this is likely contribute to Lighthouse supern peer is peering issue so they are fixing it and we'll retest it and oh lion also found Li house any other updates for White House okay not it's okay then um do we have any updates from Reon hey um yeah last week we had um of norite with of La so I didn't work too

much on P but uh about the week before uh so I'm currently prism is is only able to work with puras if the pias version is sorry the P Fork EPO is at Genesis um if you try to activate puras after Genesis there is some issue so I am working on it and also so I'm working on on P death as Fu as well um that's all for me thank you okay and anyone

from neus yeah so we're basically putting in the pias pieces inside fulu and we are pretty much just left with uh column syncing honestly and uh right now uh Nimbus is fully compliant like fully capable of uh supporting a fulu hard Fork as well so like currently whether you decide a soft fork or you decide pias at fulu as a hard Fork we should be able to work just fine and I think think at least uh two

more weeks from now I think we should be ready to form a dead yeah I mean that's F good thank you um then than yeah uh the we were preparing branch and it's 3 uh which is able to run p Fork anywhere in Electra or full and as uh there is a modood to to not switch to Electra to avoid all bugs all the net bugs we are trying to add D

supper to this Branch currently and it's a work in progress I'm not sure it's feasible but if we are able to to add the N support we will have one branch for uh then full Electra and prod oft for anywhere in the middle so it's the current work and that's so far thank you and do we have people from gring okay uh

and if not then is there anything I I missed not we could move to the defy updates um so I can just share about this status of the SP first um this is the target for the incoming release so like we were very close to hit the I mean click the release button however there are some new PR

came up and we we're now planning to uh have have been fully discussed in the ACD this week so um yeah I think the two PS are not here that's related but they mostly for for picture that's the spec side uh yeah and do we have any I mean updates or issues to discuss regarding the de dates so as I mentioned in the chat uh we currently don't support

electrogenesis so the next Dev netw workflow would be that we do a den Genesis we activate elra on uh whatever EPO like EPO one and then we would be triggering Fu on EPO OnePlus n where n is more than zero and that's where we would want to activate all the P features is that okay for every client yeah yeah I have a question so can you confirm that um the EIP 75 94 Fork Epoch will be the same than Fu Fork EP yes we can we can do that here okay so the

same at the same time yeah yeah okay perfect it's simplify simplifies my work thank you I mean ideally we can also just deprecate the EIP 7594 uh for Activation I think we should because yes uh I think uh normally we should activate P directly at Fu not after exactly exactly okay is there any opposition for just getting rid of that for Activation and just have we we needed in uh we are considering it it's it's possible soft

work we don't need for but we need for for but for how long are you going to need it for is it just for the or if we have finalized decisions that we are doing it in full I could remove it but we are talking that we are going to do it in next in a week we are going to do it in full and next we are going to to go back to to the and uh I'm not sure could we keep it for a while at least sure I I thought that the consensus was

that we're going to do this in full so that's why I'm just asking if you have strong consensus uh to do it on full okay is there any reason not to do it on full uh the talks that devet Electro devet is not stable yet and as fu is at top of Electra we could get all bus from uh from Electra net combined this box from but El is stable I would say that so like I I I feel like Electra can be considered as stable at this point

maybe not Electra devet 5 but we can probably iron out all the issues because yeah yeah I have heard the same that devet 4 is Prett stable yeah like meong is running out of the force pack and we have 99.9% participation so like we we are pretty much good so like I would be I would be comfortable using Fu as the activation trigger and prism supports it and I think also Nimo supports that yeah yeah yeah we do support it yeah what about lighthouse

Li if you are speaking we can't hear you okay you could you maybe type the message uh not we

could come back later yeah the basically the question was if it would be okay for uh Lighthouse to trigger peer death as Fu and not EIP 7594 yeah basically about demitry put in

the chat okay and maybe lotar can also uh comment now because M just joined I can repeat the question again yes please so would lar be okay with removing e 7594 at the forok activation

and just trigger P death with full uh I don't see any issue with that and G jinder's here as well he just popped in uh I can I can repeat the question one more time goer would would loer be okay to activate P death on fulu instead of EIP 7594 so that we can get rid of the EIP 7594 from the uh your config yeah I think that will be okay okay so I think there's no objection

other than from te but uh dimitry said that they can also just get rid of it so then we should probably get rid of it and lar missed the previous C updates would you like to to give a short summary of The Cent update uh CER uh yeah hello so uh we are just working on reasing uh the client uh so uh so we we basically in our in our

in the past we said that we had an issue while debugging uh the long uh the range sync uh we we figured out an issue while debugging the rain syn but turned out that uh it is not possible for us to figure out if uh a set of peers are on different Fork uh so what if we for example get blocks through one and then try to get data through the other so we are trying to figure out what is the best strategy over there uh but turns out maybe just keep picking up ear and trying out trying them out is the only reasonable strategy that we have

got uh so that way uh we we don't basically foresee any uh major change in the way we do range sync and we'll try to figure out how to best uh sort of collate those peers that we figure out that are not on a particular for uh but yeah apart from that uh nothing much has gone apart from that yeah thank you so that's go back to defate discuss bit uh any other questions or issues

regarding defate um is the make initial full spec PR going to make it to alha 10 or are we going to have off 11 for the next evet Bally this PR we were going to include it in off 9 first but it didn't make it there can you repeat the question please if you want to include this uh make initial FL spe in Alpha

10 or if you want to have a new Alpha 11 and just leave Alpha 10 as is and Target the next pet for Alpha 11 I don't have a strong opinion on that we're kind sorry go ahead no I just saying the spec sty differ to ctin if you ready then we can replace the pi and make it

merch and um especially Al the CL te okay with um EIP with IP 7594 the CER Network subnet from Dore because it's impacts P aot indeed uh yeah we should add a patch for it thank you for flagging it I I just want to comment on the time the timeline that Barnabas was asking about for for 10 or Alpha 10 or Alpha 11

at least from I don't know what you know other teams if they're as far as obviously it's December and a lot of vacation going on I don't know how active um the setting up the dev Nets or test Nets is is intended to be in the next you know we say until January um and if if it's not intended to be that active and again I don't know what the plan is then and maybe I don't I don't know if there's an urgency to alpha. 10 at that point um I I don't know who's What specifically the goal there is so we plan to have 11 five up by end of this year so for that we would need Alat 10

at least for Petra 75 but this yeah this is really appear as uh discussion but it will be based on top of that I'm not sure if we can do a p devet by the end of this year also um but the benefit is we can always have a local kosis config so if someone is working on pias over the holidays they can um otherwise they don't have to and they can reconvene in the new year yeah I don't see any big problem in merging

in the full Lo back for Alpha 10 by the way so okay yeah the

PRS that we want to discuss today maybe the same one as last time uh validator cost stuff if you want it in five or sorry in the next peret or not are you talking about the decer network subnet from Dore no the basic validator custody this one yeah I would still vote for no okay so that's going to make it into

possibly next step not yes after I guess for the basic custody um it's okay maybe I'm wrong but I think it's okay if some does implement it and some other not because it will just change the on the metadata and the not record CSC value it will just change the the value and so it will just change a count of advertis subnets for a no but if there is one node which implemented and one other node which does not should be okay then the

comments and then any other topics to discuss today any like research topic to discuss today no okay anything else before

we closing this code no no just minor question during the during the Death code we really briefly talked about possible pure death uh interrupt um I don't know if it was a serious discussion or not um

and is it mean in person yeah yeah yeah Qui quick informal discussion about about that do you mean the usual um R&D interrupt the same one we had for Petra or do you mean a specific uh off band One um specific for p because uh yes specific for p we really briefly talked it I mean specific for p would be specific for Fu at this point right yeah yeah yeah so everybody who's going to be working on the next Fork which is basically everyone would be there so at this point

that would be just normal interrupt I think that's that's right that's right I think it would have only made sense to do it maybe this year to like actually push a separate parallel session but it feels like no team had enough bandwidth to do a proper parallel uh line next Spectra so I I think uh parallel implementation is actually not working that well yes any other comments on

the interrupt or like here Des R this not we can wrap up earlier thank you for attending have a nice day thank you bye see you

guys yeah he