pull down to refresh

I stumbled upon this comment from Shinobi which is quite interesting.
I will repeat this for people who weren't paying attention: Knots 0.29 by default will NOT relay the transaction types that all Lightning implementations are working to implement for non-cooperative closes.
Nitesh B did a further explanation:
For those curious on what Shinobi is talking about. Ephemeral anchors or Ephemeral dust are basically special type of Bitcoin transactions that can be relayed without any fee associated with them with the condition that they are relayed as part of a "package" which has a child transaction which pays enough fees for the whole thing. What this allows is anyone can use that output as an input to a child transaction which helps create a fee paying child transaction. This is great in lightning because fee estimation sucks because you sign transactions with your peer well ahead of time and you have no idea what the fee rate is going to be in the future. Now anyone can use an ephemeral output as the input to a child tx which can help add fees at the time of broadcasting the tx. Previously we got around this problem by adding 2 anchor outputs to the commitment transaction and nodes just put sufficient fees hoping the tx gets relayed. But now because we have package relay we can now fee bump peacefully even if the fees attached to the commitment tx is below the min relay fee rate of the bitcoin nodes.
Obviously you can change the default.
21 sats \ 7 replies \ @ek 3h
Not saying Shinobi is wrong or right, just want to remind everyone to also not trust arguments from "your side" without some due diligence. If you want to use someone else's arguments, make sure you actually understand them first, for example by using your own words.
we really need to add twitter embeds for xcancel links, too
reply
we really need to add twitter embeds for xcancel links, too
yeaah! :)
reply
100 sats \ 0 replies \ @ek 2h
/cc @Scoresby, you also asked for this iirc
mhh, but my bot (@nitter) will not reply with xcancel.com links to xcancel.com links
reply
0 sats \ 4 replies \ @ek 3h
oof, what a horrible take this is though:
reply
yes. I really doubt that most of Knots (and core) node runner ever change any default configuration value. And this is also a reason why it's kind of sketchy what core is doing.
reply
10 sats \ 0 replies \ @ek 2h
Yeah I also doubt most runners change defaults. Defaults matter.
a reason why it's kind of sketchy what core is doing.
@bluematt mentioned the following in this debate (starts at 10:09):
Is it a requirement for software engineers to maintain a feature that is strictly harmful to your local node? It does not prevent you from storing the same data, it's going to end up on your disk, it's going to end up in your p2p connection one way or another, either because it's in a block or before then. [...]
I think it makes sense.
However, he mentions maintenance though, but I think no code was removed, only defaults changed. But maybe he meant the future when the option including the code is removed, if ever.
reply
I think both were gaslightning with that tweet.
reply
21 sats \ 3 replies \ @DarthCoin 2h
Already discussed yesterday here: #1222299
reply
oh, thanks. Did miss that one.
reply
21 sats \ 1 reply \ @DarthCoin 2h
😂😂😂
reply
🤣 done!
reply
What's the default setting for Knots to relay those transactions?
reply
I think these settings are related: -permitephemeral (which is "anchor" in knots) -mempooltruc
reply
I did a quick search but couldn’t find a page that explains all the settings. Do you know if there’s one that shows and explains everything?
reply
reply
I didn’t see those parameters, am I missing something?
reply
permitephemeral does not exist in Bitcoin Core
reply
Oops! In Knots, the '-permitephemeral' parameter has three options (send, dust, and anchor). If we wanna enable all three, should the file look like this?
-permitephemeral=send -permitephemeral=dust -permitephemeral=anchor
Would be nice if there was a site that broke down every parameter properly.
reply