`ChannelMonitorUpdate`s are our most size-sensitive objects - they
are the minimal objects which need to be written to disk on each
commitment update. Thus, we should be careful to ensure we don't
pack too much extraneous information into each one.
Here we add future support for removing the per-HTLC explicit
`Option<Signature>` and `HTLCInCommitmentUpdate` for non-dust HTLCs
in holder commitment tx updates, which are redundant with the
`HolderCommitmentTransaction`.
While we cannot remove them entirely as previous versions rely on
them, adding support for filling in the in-memory structures from
the redundant fields will let us remove them in a future version.
We also add test-only generation logic to test the new derivation.
This is largely motivated by some follow-up work for anchors that will
introduce an event handler for `BumpTransaction` events, which we can
now include in this new top-level `events` module.
There is no need to fill the user's logs with errors that are expected
to be hit based on specific edge cases, like providing preimages after
a monitor has seen a confirmed commitment on-chain.
This doesn't really change our behavior – we still apply and persist the
state changes resulting from processing these updates regardless of
whether they succeed or not.
This results in a new, potentially redundant, `ChannelMonitorUpdate`
that must be applied to `ChannelMonitor`s to broadcast the holder's
latest commitment transaction.
This is a behavior change for anchor channels since their commitments
may require additional fees to be attached through a child anchor
transaction. Recall that anchor transactions are only generated by the
event consumer after processing a `BumpTransactionEvent::ChannelClose`
event, which is yielded after applying a
`ChannelMonitorUpdateStep::ChannelForceClosed` monitor update. Assuming
the node operator is not watching the mempool to generate these anchor
transactions without LDK, an anchor channel which we had to fail when
deserializing our `ChannelManager` would have its commitment transaction
broadcast by itself, potentially exposing the node operator to loss of
funds if the commitment transaction's fee is not enough to be accepted
into the network's mempools.
Currently, all that is required to force close a channel is to broadcast
either of the available commitment transactions, but this changes with
anchor outputs – commitment transactions may need to have
additional fees attached in order to confirm in a timely manner. While
we may be able to just queue a new update using the channel's next
available update ID, this may result in a violation of the
`ChannelMonitor` API (each update ID must strictly increase by 1) if the
channel had updates that were persisted by its `ChannelMonitor`, but not
the `ChannelManager`. Therefore, we choose to re-purpose the existing
`CLOSED_CHANNEL_UPDATE_ID` update ID to also apply to
`ChannelMonitorUpdate`s that will force close their respective channel
by broadcasting the holder's latest commitment transaction.
...replacing it with an acessor `addresses()`.
Besides removing a redundant data structure already present on inner
`NodeAnnouncement`, this change makes it possible to discover new address types
upon deserialization thanks to `UnsignedNodeAnnouncement`'s implementation.
`vec_type` is confusing - it is happy to have a missing entry,
"reading" an empty `Vec` instead, but always writes something,
making a serialization round-trip different.
This is a problem for writing a new `Vec` which is
backwards-incompatible, but only if filled in. In that case we'd
really like the same read behavior, but not write anything if the
`Vec` is empty. Here we introduce such semantics via a new
`optional_vec` TLV format.
Since the claim events are stored internally within a HashMap, they will
be yielded in a random order once dispatched. Claim events may be
invalidated if a conflicting claim has confirmed on-chain and we need to
generate a new claim event; the randomized order could result in the
new claim event being handled prior to the previous. To maintain the
order in which the claim events are generated, we track them in a Vec
instead and ensure only one instance of a PackageId only ever exists
within it.
This would have certain performance implications, but since we're
bounded by the total number of HTLCs in a commitment anyway, we're
comfortable with taking the cost.
The previous documentation was slightly incorrect, a `Claim` can also be
from the counterparty if they happened to claim the same exact set of
outputs as a claiming transaction we generated.
Since we don't store `pending_claim_events` within `OnchainTxHandler` as
they'll be regenerated on restarts, we opt to implement `PartialEq`
manually such that the field is not longer considered.
`serde` doesn't bother with MSRVs, so its expected to break
frequently. Yesterday, the `derive` feature had its MSRV broken in
a patch version without care.
Luckily its trivial for us to remove the `serde` dependency in
`lightning-block-sync`, using only `serde_json` for the JSON
deserialization part. It even ends up net-negative on LoC.
If we have a public channel which doesn't yet have six
confirmations the network can't possibly know about it as we cannot
have announced it yet. However, because we refuse to include
route-hints if we have any public channels, we will generate
invoices that no one can pay.
Thus, if we have any public, not-yet-announced channels, include
them as a route-hint.
1.48.0 was released at the end of 2020, nearly 2.5 years ago. It
has been the rustc available on Debian stable since bullseye,
released in 2021. supporting Debian oldstable for more than a year
seems more than sufficient time to give Debian folks to upgrade,
and bullseye is set to become `oldstable` later this year with the
release of `bookworm`, likely this summer.
This also allows us to clean up our MSRV substantially, having a
single MSRV across our crates rather than a number of separate
ones. Sadly, windows already requires 1.49.
