-
Notifications
You must be signed in to change notification settings - Fork 179
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
BIP78 PayJoin to .onion fails, fallback tx is not broadcast after timeout #635
Comments
Next attempt! @openoms torified his JM+Core, I started a new ephemeral hidden service. JM decoded the URI properly. JM successfully setup a connection to the P2EP. [I got notification about the negotiation in Wasabi] Wasabi sent a wrong PSBT [probably because I am running an out of date branch, with last development in April, and the BIP has changed since?]
However again, no fallback transaction is broadcast by JM.
|
Would like to add to here: Should be able to communicate the PayJoin over Tor without needing to torify the bitcoindRPC calls too. |
Right, a couple of things here. (1) fallback on client side .. it is not implemented properly yet; my bad. I will do this shortly. It's pretty simple (just a simple (2) hidden service receiver - that is also not implemented at all yet. If anyone feels inclined to work on it, I can suggest you look at Edit: oh and btw thanks very much for testing guys, that is hugely useful. |
Another test with https://demo.payjoin.kukks.org Here the PJ does not happen repeatedly with the error:
|
I used https://btc.donate.kukks.org/ a couple of days ago with a similarly small amount and it worked fine for me. You seem to have a successful payment given the service was unavailable, as it says ... probably they are out of suitable utxos to serve? (it is actually good and correct that we don't easily understand what caused the error .. the server should not be telling everything about its internal state). |
It's really really difficult to see you guys working on such awesome magic, and then keep the cool and not prematurely test everything... 😀 Since these reported issues are expected, feel free to close the issue to keep the repo clean; or close it after it's implemented, up to you!
afaik, @Kukks has plenty of coins on that server. |
No seems fine here, good a place as any to keep track of these updates.
Oh I just remembered. the problem is simple - you need bech32. As explained elsewhere (maybe #536 I forget where i wrote it), btcpayserver plans to have support for merchant wallets to do p2sh wrap and bech32 both in the same wallet/merchant endpoint, but now they don't have it. |
@waxwing can confirm that PJ works with a
Looking great: https://blockstream.info/testnet/tx/d57f86a2435a08c8b7ddca04cb1dbf1a522b869139a48afba427bb9ff119d2dd |
Wow, this is great. About the old-discarded Wasabi PoC branch withpayjoin support: do not use it. It was created to show an idea and the implementation is previous to the bip78 discussions (it doesn't support none of these: disableoutputsubstitution, additionalfeeoutputindexm maxadditionalfeecontribution, minfeerate) and it returns transactions encoded as hex only. |
Changing title slightly to better reflect situation and will keep it open until the hidden service stuff is done client-side. |
Added support for paying to hidden service in this branch: https://github.com/JoinMarket-Org/joinmarket-clientserver/tree/hiddenservice-bip78 (see #638 ). If you guys could test it that'd be great; it worked for me with the ephemeral HS i set up. I was using p2sh wrap wallets both sides, and a v2 onion; I think it should work fine with native wallet and v3 onion, but that'd be of interest to test (well; just testing any combination is super interesting, if we can get a Joinmarket to Wasabi payment work). |
I have tested a PayJoin with latest master between two bech32 JM wallets. RPC connection over Tor.
Sender output: Did not have the opportunity test to test with a Tor HS endpoint yet. |
Just to be clear: the above experiment is not connected with the topic of this thread, which is BIP78 Payjoin between I won't remove/deprecate the old JM-JM version until at least we have a BIP78 receiver implemented, which could be done soonish if anyone had the intent to do it (see the test/payjoinserver code). But this is already seeming to cause a lot of confusion. Thanks for the bug report and the positive test result. That bug is easy to fix (and indeed, harmless). |
@openoms and I are testing cross-wallet PayJoin functionality.
Setup
Openoms is sending from a non-torified JoinMarket wallet master branch.
I am receiving in Wasabi Wallet to an ephemeral hidden service. [it's a reckless feature, not planned to be released, this is mainly to test JM, not WW.]
The URI got properly decoded, but the hostname lookup failed with no result.
Problem
I would say, that in such a case, the Joinmarket wallet should fall back to the non-PayJoin transaction, and broadcast this directly.
Joinmarket Logs
The text was updated successfully, but these errors were encountered: