This is a helper function that we will need to accurately determine the
weight of inputs specified in a PSBT.
Due to the nature of P2WSH and script-spend P2TR inputs, we can only
accurately estimate their weights if the full witness is already known.
So this helper function rejects inputs that use a script spend path but
don't fully specify the complete witness stack.
We want to re-use the logic that determines what change amount is left
over depending on whether we add or don't add a change output to a
transaction, respecting the change output's dust limit.
We'll want to be able to tell the coin selection algorithm that we
intend to add any change to an existing output instead of assuming that
a change output is always created.
If we add any left over change to an existing output, we can skip the
dust amount check as we assume the selected existing output already has
a non-dust amount requested (responsibility of the caller to assert).
Before this commit the coin selection logic in the chanfunding package
would always assume that there is a P2WSH funding output and potentially
a P2TR change output. But because we want to re-use the coin selection
for things other than just channel funding, we make the logic more
generic by allowing us to specify both the existing weight of the
transaction (the already known, static parts of the TX) as well as the
type of the potential change output we would use.
With this commit we prepare for the lnwallet channel funding logic to be
aware of the config-level coin selection strategy by adding it to the
wallet config.
In this commit, we start persisting shutdown info when we send the
Shutdown message. When starting up a link, we also first check if we
have previously persisted Shutdown info and if we have, we start the
link in shutdown mode meaning that it will not accept any new outgoing
HTLC additions and it will queue the shutdown message after any pending
CommitSig has been sent.
ShutdownInfo contains any info about a previous Shutdown message that we
have sent. This commit adds this type along with read and write methods
for it in the channel db. The existence of the ShutdownInfo on disk
represents the fact that we have previously sent the Shutdown message
and hence that we should resend it on re-establish.
Over the last few commits we have systematically eliminated all but
two states. This allows us to replace it with a boolean to encode
the two remaining states. We would like to be able to eliminate this
field entirely, but doing so requires being able to prove that the
concurrent request block is necessary. This is more difficult and
will be left to future commits.
lnwallet: remove unused channelPendingPayment channelState
Since this state is never set nor read, we remove it completely.
lnwallet: remove redundant channelDispute channelState
In this case, even though we do set this value, it is never read.
Further, the times we read the field at all from LightningChannel
we want the situation of force-closure to block any other
concurrent closure attempts, so we change the sites where we set
channelDispute to channelClosed.
lnwallet: remove redundant channelClosing channelStatus
This value is never used to impact control flow so we need not set
it. We also need not have it.
bitcoind v25.0 updated the `sendrawtransaction` RPC to have an optional
argument `maxburnamount` with a default value of 0. This means our
existing test that uses burning output cannot be published, hence, we
remove the usage of it in our tests and replace it with a normal tx.
Previously, `PublishTransaction` in `btcwallet` will first mark the tx
label in db first before broadcasting it to the network. Even though the
broadcast may fail with the error already in mempool or already
confirmed, this tx label updating is still performed. To maintain the
old behavior, we now catch the errors from `TestMempoolAccept`, and make
sure to call the `PublishTransaction` to mark the tx label even there
are errors from mempool acceptance check.
This commit adds a mempool acceptance check before broadcasting a given
transaction. To maintain the current behavior from
`BtcWallet.PublishTransaction`, the two errors, `ErrInMempool` and
`ErrAlreadyConfirmed` returned from `TestMempoolAccept` are ignored.
We take into account a fee buffer of twice the current fee rate
of the commitment transaction plus an additional htlc output
when we are the opener of the channel hence pay when publishing the
commitment transaction. This buffer is not consensus critical
because we only consider it when we are in control of adding a
new htlc to the state. The goal is to prevent situations
where we push our local balance below our channel reserve due to
parallel adding of htlcs to the state. Its not a panacea for these
situations but until we have __option_simplified_update__ deployed
widely on the network its a good precaution to protect against
fee spikes and parallel adding of htlcs to the update log.
Moreover the way the available balance for a channel changed.
We now need to account for a fee buffer when we are the channel
opener. Therefore all the tests had to be adopted.
Prior to this commit, taproot channels had a bug:
- If a disconnect happened before peer.AddNewChannel was called,
then the subsequent reconnect would call peer.AddNewChannel and
attempt the ChannelReestablish dance.
- peer.AddNewChannel would call NewLightningChannel with
populated nonce ChannelOpts. This in turn would call
InitRemoteMusigNonces which would create a new musig pair session
and set the channel's pendingVerificationNonce to nil.
- During the reestablish dance, ProcessChanSyncMsg would be called.
This would also call InitRemoteMusigNonces, except it would fail
since pendingVerificationNonce was set to nil in the previous
invocation.
To fix this, we add a new functional option to signal to the init logic
that it doesn't need to call InitRemoteMusigNonces in in
ProcessChanSyncMsg.
