force closed channel with fee too low

[nicklywhoop]

Hi,

I am trying to recover my sats from my forced closed channel, I have searched for quite a bit but everything I tried isn't working.
I have a closing transaction with 17s/vb which I am trying to fee bump without success

https://mempool.space/tx/309717fe30e149df75be1f7059ddce1feec8ebe3e8ca5e09fb636971269f8f33

Commands I have tried

lncli --rpcserver=lnd.embassy wallet bumpclosefee --conf_target 1 5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1

No output (exit 0) but no CPFP transaction created

lncli --rpcserver=lnd.embassy wallet bumpfee --conf_target 1 --force 309717fe30e149df75be1f7059ddce1feec8ebe3e8ca5e09fb636971269f8f33:0

[lncli] rpc error: code = Unknown desc = the passed output does not belong to the wallet

chantools sweepremoteclosed --recoverywindow 30 --feerate 150 --sweepaddr <> --rootkey <> --publish

{"code":-26,"message":"non-BIP68-final"}

Please help :)

[guggero]

This is an anchor channel, so it should be possible to CPFP it. But only lnd will do that, chantools doesn't have that functionality implemented. Does the channel show in lncli pendingchannels? If yes, then lnd should take care of it (assuming the the balance is yours?). If not, then how exactly was the channel force closed?

[nicklywhoop]

When restoring the backup string, the log doesn't give a lot of information and nothing seems to be failing

2023-11-28T05:44:06+00:00  2023-11-28 05:44:06.326 [INF] CHBU: Restoring ChannelPoint(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1) to disk:
2023-11-28T05:44:06+00:00  2023-11-28 05:44:06.380 [INF] LTND: SCB Recovery: created channel shell for ChannelPoint(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1), chan_type=42
2023-11-28T05:44:06+00:00  2023-11-28 05:44:06.380 [INF] LTND: Inserting 1 SCB channel shells into DB

On the other hand if I try to run a restorechanbackup I get the following output:

2023-11-28 09:35:01.398 [WRN] PEER: Peer(030a58b8653d32b99200a2334cfe913e51dc7d155aa0116c176657a4f1722677a3): Unable to find our forwarding policy for channel 5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1, using default values
2023-11-28 09:35:01.398 [INF] HSWC: ChannelLink(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1): starting
2023-11-28 09:35:01.398 [INF] CNCT: Attempting to update ContractSignals for ChannelPoint(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1)
2023-11-28 09:35:01.398 [INF] HSWC: ChannelLink(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1): HTLC manager started, bandwidth=0 mSAT
2023-11-28 09:35:01.398 [INF] HSWC: ChannelLink(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1): attempting to re-synchronize
2023-11-28 09:35:01.399 [INF] HSWC: ChannelLink(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1): received re-establishment message from remote side
2023-11-28 09:35:01.399 [ERR] LNWL: ChannelPoint(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1): sync failed with local data loss: remote believes our tail height is 1, while we have 0!
2023-11-28 09:35:01.399 [WRN] LNWL: ChannelPoint(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1): detected restored triggering DLP
2023-11-28 09:35:01.399 [WRN] HSWC: ChannelLink(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1): error when syncing channel states: ChannelPoint(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1) with CommitPoint(03bf21f222bb573685a32198f61ef21ef54d8290cb81f68a7796b4666be7d32120) had possible local commitment state data loss
2023-11-28 09:35:01.400 [ERR] HSWC: ChannelLink(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1): failing link: unable to synchronize channel states: ChannelPoint(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1) with CommitPoint(03bf21f222bb573685a32198f61ef21ef54d8290cb81f68a7796b4666be7d32120) had possible local commitment state data loss with error: unable to resume channel, recovery required
2023-11-28 09:35:01.400 [INF] HSWC: ChannelLink(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1): exited
2023-11-28 09:35:01.401 [INF] HSWC: ChannelLink(5da31680c15fe3d899bf4ec8823797d9b74acdc46edb38384b1cee737862f0cd:1): stopping

That's a lot of concepts I don't understand if you can help 🤞

[guggero]

Thanks for the log. That looks as I would expect.
But I did some digging into the code, and it looks like you hit an edge case here: In case of recovering a channel with anchor outputs, your node relies on the other node to CPFP the transaction, using their anchor. But since all the balance is on your side, only your node has an anchor to pull down the TX with CPFP. And normally your node would do that, if the channel was still in the normal, active state.
But since you recovered the channel, lnd is not attempting the CPFP by itself, since the channel is in the "waiting close" state.

So I think you have two options here:

1. Just wait for the mempool to clear somewhat and the TX confirming by itself.
2. Wait for someone (most likely me) to implement a new sweepanchor functionality into chantools.

[nicklywhoop]

Great, I think that make sense to me why it is an edge case and happy to wait I initially thought I must be running the wrong command.
Hopefully the mempool clears soon or you (or someone else) find some time to add the functionality to chantools.
Thank you so much for your help on this :)

[guggero]

@nicklywhoop you can now try out the pull request I opened for doing a manual fee bump of an anchor output: lightninglabs/chantools#100

[nicklywhoop]

Wow it worked! I got my funds back 👍
Thank you so much!