Ethereum Protocol Fellowship (EPF) Cohort 7 — Applications open until May 13

FOCIL Breakout #019

2025-09-09 Agenda: #1720 canonical JSON

Transcript

00:04:40
Jihoon:We have a small set of people here.
00:04:45
Jihoon:I guess people are probably busy with Fusaka, so we can keep this one short.
00:04:51
Jihoon:I don't have a special agenda other than development updates, so…
00:04:56
Jihoon:We can jump into it first, and see if anything else comes up.
00:05:00
Jihoon:Firstly, I want to give a heads up about the spec. Currently, we cannot see the fossil spec on the website.
00:05:07
Jihoon:I was told that feature specs were hidden because they are experiments, but shorter forks require some people to work on the next fork, so I asked Justin about this, and
00:05:19
Jihoon:I think feature specs would have a separate section from canonical ones.
00:05:24
Jihoon:As far as I know, Justin is on his vacation, so we can wait a little bit more.
00:05:30
Jihoon:Next up would be rebasing. We are rebasing onto Fulu and then Coloss. Coloas is in no fork, like, without a PBS or whatnot.
00:05:44
Jihoon:I know that not every client teams are here, but
00:05:49
Jihoon:So we are talking about rebasing, so does any client that want to share some updates?
00:06:10
Jihoon:Maybe not. I think people are still busy with Fusaka, so, I think we can…
00:06:16
Jihoon:resume our work on Fossil after…
00:06:20
Jihoon:I don't know, shipping full circle or whatever, so that, so we can try out client interrupts again after we settle on
00:06:29
Jihoon:I guess, clue us with the updated engine APIs.
00:06:36
Jihoon:maybe we have the first EL interrupt. I haven't tried it out yet, but perhaps, Bella, you wanna share some updates?
00:06:49
Jihoon:I… guess you reached the EL interop between RET and GATH.
00:06:58
Pelle Krabbenhoeft:Oh yeah, I was just doing some basic testing on Electra, and everything seems to be working so far. I was only able to run it for a short amount of time yesterday.
00:07:10
Jihoon:Great, you might want to double-check if
00:07:13
Jihoon:Red produce IELTS correctly. You can see the log, or click some slot in DORA to see the IELT tab.
00:07:22
Pelle Krabbenhoeft:Okay, I can take a double check at that.
00:07:27
Jihoon:So, I guess we have… We already have 6 clients ready for interop.
00:07:36
Jihoon:Do we have any update on the metrics, Katya?
00:07:45
Katya Riazantseva:Hello, everyone.
00:07:47
Katya Riazantseva:I've just added a couple of metrics more, but I think they are my articles, so I just build them.
00:07:55
Katya Riazantseva:Probably next week, I will be more… I will have more free time, and maybe we will discuss.
00:08:04
Katya Riazantseva:the metrics, so I can open the APR in the metrics repo, if we have some agreements on metrics, if you don't mind, Jehoon.
00:08:16
Jihoon:Yeah, of course, yeah, yeah, feel free to share it.
00:08:20
Katya Riazantseva:And I wanted…
00:08:22
Katya Riazantseva:Yeah, and I wanted to ask about the,
00:08:26
Katya Riazantseva:the config, the current config, so we are, running…
00:08:33
Katya Riazantseva:fossil on Fuller, on top of full, right? Or maybe glass already, or what's the current status?
00:08:41
Jihoon:So, we… were based on Electra, but now that it's
00:08:48
Jihoon:quite stale, so I think only Lighthouse and Geth are working at the moment. I repaired a new Cortosis config and Docker image that runs on top of Electra. It runs with only Lighthouse and Geth pair.
00:09:06
Jihoon:So maybe you can try that out.
00:09:10
Katya Riazantseva:Because Lone Star is already rebased on Fulu, also, so yeah, I was trying to… Combine it with guest.
00:09:21
Katya Riazantseva:It didn't work properly, I don't know, maybe it's CL side, we'll check again.
00:09:29
Jihoon:Oh, okay, so I… okay, I… let me try that interop between low star and get on top of…
00:09:36
Jihoon:It started with the fluid Genesis work, right?
00:09:39
Katya Riazantseva:No, no. I actually, I failed with Foologenesis. I'm starting from Electra, then follow, then go to Fossil.
00:09:49
Jihoon:Correct. Can I already start from Pulu?
00:09:55
Jihoon:Ethereum package supports Flu… and it also supports Coloss Genesis fork, so you should be able to run with Fluid Genesis.
00:10:04
Jihoon:But yeah, I… yeah, I saw your, some error message in the, in the Discord channel, but…
00:10:10
Jihoon:Yeah, I'm not sure what's the issue.
00:10:13
Katya Riazantseva:Yeah, I will… I will give it a try again soon.
00:10:18
Pelle Krabbenhoeft:I also tried Lodestar and Gep for the Fulu Fork, and Fulu was fine, and then as soon as, Fossil Fork started, everything went… not… it all stopped working.
00:10:31
Pelle Krabbenhoeft:I couldn't quite figure it out, either.
00:10:35
Katya Riazantseva:Yeah, I've seen the reason, it's like,
00:10:39
Katya Riazantseva:on the, I guess, on the Lodestar, so probably I will ask,
00:10:45
Katya Riazantseva:and C, because, NC is here. So, the reason was in the,
00:10:52
Katya Riazantseva:engine API, I guess so,
00:10:57
Katya Riazantseva:Update, how does call update payload, or…
00:11:00
Katya Riazantseva:Notified payload, or something like this.
00:11:04
Katya Riazantseva:Yeah, this is where it failed last time I ran it.
00:11:12
Jihoon:Yeah, I can try that out. Maybe I have to update, some GAT code regarding the engine API version. Maybe, Nancy, do you have any clue regarding this?
00:11:27
NC:to be honest, like, the new code, I haven't ran it at all. It's not tested at all, so…
00:11:36
Jihoon:Right, so there was no change on low star, so yeah, maybe it's natural for…
00:11:42
Jihoon:Low start to fail after we transition into flu.
00:11:50
Katya Riazantseva:So, it happens right after transitioning to fossil, to EIP, like, from full to EIP. So,
00:11:58
Katya Riazantseva:Yeah, I will take a deep look into it.
00:12:03
Jihoon:Okay, so, are you trying that…
00:12:08
Jihoon:I mean, I retry EIP7805 after Fallout, because…
00:12:14
Jihoon:Because we are rebasing, right?
00:12:20
Jihoon:But I guess a low star is…
00:12:23
Jihoon:they don't have any update on their rebasing yet, so maybe you can just start on Electra and then IP7 and 05, and see how your metrics work.
00:12:34
Katya Riazantseva:No, we actually rebased, right? And see, we are on top of who currently is.
00:12:39
Jihoon:Okay, so we were talking about rebasing, okay.
00:12:44
Jihoon:Right, yeah, I can try that as well to see what's the log, and see…
00:12:52
Jihoon:And then I can share the, what I found.
00:13:05
Katya Riazantseva:And the latest, clients, images, so they are still in your repo, right? Updated?
00:13:16
Katya Riazantseva:the configs, etc.
00:13:19
Jihoon:Which… which image are you asking? I guess?
00:13:25
Katya Riazantseva:I normally… I'm using your repo with all the configs of the clients, so…
00:13:30
Jihoon:Yeah, so, we have two versions. The latest version is…
00:13:37
Jihoon:The latest one reflects the updated engine API, and the one before the latest one is previous one. So, I guess, if you are playing with Blowestar, then they haven't made any change yet, so you can just use the, the old one.
00:13:57
Katya Riazantseva:Okay, I can try both.
00:14:04
Jihoon:Let me pull that tag real quick, so that… Yeah, so you can try…
00:14:14
Jihoon:Oh, there's the chat? Yes.
00:14:19
Jihoon:Yeah, this one should be good.
00:14:29
Jihoon:Does anyone have anything to discuss?
00:14:38
Katya Riazantseva:So currently, people are working on EPBS, right?
00:14:43
Katya Riazantseva:And, everyone's busy with PPS and Osaka at the same… at the same time.
00:14:52
Jihoon:I… think so. I think people are usually…
00:14:55
Jihoon:Most of the people are working on Fusaka, and some people are working on the PBS, on the CL side, at least, and some people
00:15:05
Jihoon:Find some time to work on next fork.
00:15:09
Jihoon:Like, 6-second slots, for example.
00:15:28
Jihoon:Yeah, we can… if anyone has anything to discuss, then we can, cover it. Otherwise, we can just…
00:15:54
Jihoon:Right, guys, I think this is it. Thanks for joining the call. Hope you guys have a great rest of your day, and…
00:16:04
Katya Riazantseva:Thank you, Bargain.

Chat Logs

00:08:20
Jihoon:https://hackmd.io/2W-xw95aTOOMtDuObFTGRQ

Summary

7 highlights · 3 action itemsExperimental

spec and rebasing

  • Feature specs hidden on website; Justin working on separate section for experimental forks00:05:07
  • Rebasing onto Fulu, then Coloss (without PBS)00:05:42

client updates

  • First EL interop working: Reth and Geth on Electra00:07:07
  • 6 clients ready for interop00:11:41
  • Lodestar rebased on Fulu; Lighthouse/Geth still on Electra00:12:26

testing progress

  • Lodestar-Geth pairing fails at Fulu→FOCIL transition due to engine API issue00:10:31
  • New Coloss config/Docker image available; runs Lighthouse-Geth only on Electra base00:12:51

Action Items

  • Pelle (Reth team): Verify Reth produces IELTs correctly via logs or DORA00:06:26
  • Jihoon and Katya: Investigate Lodestar-Geth engine API failure during Fulu→FOCIL transition00:12:51
  • Katya: Open PR in metrics repo after team discussion on metric agreements00:08:03

Targets

  • Next FOCIL call in 2 weeks (2025-09-23)00:16:01