[bug]: force-closure in a simple-taproot channel between LND and Zeus

[starius]

Background

Simple taproot channel was force closed when it was not used. I don't know why.

Your environment

• version of lnd: v0.18.2-beta-3-gf5378c239 (my own build of v0.18.2 + sweep, rpcserver: optimize SendAll cost wise #8936 )
• which operating system (uname -a on *Nix): Linux vps12345 6.1.0-13-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.55-1 (2023-09-29) x86_64 GNU/Linux
• version of btcd, bitcoind, or other backend: bitcoin 27.0

My peer who opened a simple taproot channel to me. is using Zeus v0.9.0 iOs

Steps to reproduce

My peer opened a simple-taproot channel to me (LND) from Zeus v0.9.0 iOs. He paid an invoice (so my node forwarded it). Everything worked well. However in several hours he approached me and told that the channel is force closed. I inspected the logs (see below) and found that my node (LND) initiated the force-closure.

Expected behaviour

I expected the channel not to be force-closed.

Actual behaviour

The channel was force-closed by my node.

Notes

Part of the log (obfuscated sensitive information like channel_point, capacity, SCID etc):

2024-09-12 22:07:19.659 [INF] PEER: Peer(peer_id): Loading ChannelPoint(channel_point), isPending=false
2024-09-12 22:07:19.667 [INF] HSWC: ChannelLink(channel_point): starting
2024-09-12 22:07:19.667 [INF] HSWC: Trimming open circuits for chan_id=SCID, start_htlc_id=0
2024-09-12 22:07:19.667 [INF] HSWC: Adding live link chan_id=some_other_identifier, short_chan_id=SCID
2024-09-12 22:07:19.667 [INF] CNCT: Attempting to update ContractSignals for ChannelPoint(channel_point)
2024-09-12 22:07:19.667 [INF] HSWC: ChannelLink(channel_point): HTLC manager started, bandwidth=bandwidth mSAT
2024-09-12 22:07:19.667 [INF] HSWC: ChannelLink(channel_point): Attempting to re-synchronize channel: SCID=SCID, status=ChanStatusDefault, initiator=false, pending=false, local commitment has height=6, local_htlc_index=0, local_log_index=3, remote_htlc_index=3, remote_log_index=3, remote commitment has height=6, local_htlc_index=0, local_log_index=3, remote_htlc_index=3, remote_log_index=3
2024-09-12 22:07:19.668 [INF] PEER: Peer(peer_id): Negotiated chan series queries
2024-09-12 22:07:19.668 [INF] DISC: Creating new GossipSyncer for peer=peer_id
2024-09-12 22:07:19.745 [INF] DISC: GossipSyncer(peer_id): filtering chan range: start_height=0, num_blocks=861043
2024-09-12 22:07:19.746 [INF] HSWC: ChannelLink(channel_point): received re-establishment message from remote side
2024-09-12 22:07:19.862 [ERR] HSWC: ChannelLink(channel_point): failing link: ChannelPoint(channel_point): unable to accept new commitment: no partial signature found with error: invalid commitment
2024-09-12 22:07:19.916 [ERR] HSWC: ChannelLink(channel_point): link failed, exiting htlcManager
2024-09-12 22:07:19.916 [INF] HSWC: ChannelLink(channel_point): exited
2024-09-12 22:07:19.971 [INF] HSWC: ChannelLink(channel_point): stopping
2024-09-12 22:07:19.971 [INF] HSWC: Removing channel link with ChannelID(some_other_identifier)
2024-09-12 22:07:19.971 [WRN] PEER: Peer(peer_id): Force closing link(SCID)
2024-09-12 22:07:19.971 [INF] CNCT: Attempting to force close ChannelPoint(channel_point)

When my peer was opening the channel, he succeeded only after disabling Zeus feature "attempt to use SCID pseudonym". Maybe this is important.

[Roasbeef]

2024-09-12 22:07:19.862 [ERR] HSWC: ChannelLink(channel_point): failing link: ChannelPoint(channel_point): unable to accept new commitment: no partial signature found with error: invalid commitment

This was fixed by #9046 (included in 0.18.3).