Antoine Riard [Tue, 28 Apr 2020 00:07:17 +0000 (20:07 -0400)]
Dedup RemoteTxCache by removing OnchainTxHandler copy
RemoteTxCache was providing all data needed at transaction
signature for any remote HTLC transaction or justice transaction.
This move was making the API between OnchainTxHandle akward and
scope of responsibilites with ChannelMonitor unclear.
Instead scope OnchainTxHandler to transaction-finalization, fee-bumping
and broadcast only.
Antoine Riard [Wed, 6 May 2020 00:00:01 +0000 (20:00 -0400)]
Extend KeysInterface with derive_channel_keys
A dynamic-p2wsh-output like `to_local` on local commitment/HTLC txn
require a signature from delayed_payment_key to be spend. Instead of
sending private key in descriptor, we ask for spender to derive again
the corresponding ChannelKeys based on key state, uniquely identifying
a channel and encompassing its unique start data.
Antoine Riard [Tue, 12 May 2020 00:29:39 +0000 (20:29 -0400)]
Add KeysManager::derive_unique_start
In next commit, we add derive_channel_keys, a helper to rederive
keys from a SpendableOuptutDescriptor. We still need to ensure
that user rederive same set of keys even if KeyManager is started
at a later time. For doing so we need to pass starting time as
part of the re-derived sha_state, therefore we need to make it
accessible and part of channel state key material.
Seed is already required to be conserved between KeysManager
serializations.
Antoine Riard [Tue, 24 Mar 2020 03:28:00 +0000 (23:28 -0400)]
Replace is_htlc in InputMaterial by InputDescriptor
As we cache more and more transaction elements in OnchainTxHandler
we should dry up completly InputMaterial until them being replaced
directly by InputDescriptor
As we can't predict if any and which revoked commitment tx is
going to appear onchain we have by design to cache all htlc information
to regenerate htlc script if needed.
This caused a bunch of cascading changes, including
passing loggers down to Channels in function calls
rather than having each Channel have a pointer to the
ChannelManager's Logger (which was a circular reference).
Other structs that the Channel had passed its Logger to also
had their loggers removed. Other newly unused Loggers were
also removed, especially when keeping them would've caused
a bunch of extra test changes to be necessary, e.g. with
the ChainWatchInterfaceUtil's Logger.
Matt Corallo [Tue, 12 May 2020 01:09:44 +0000 (21:09 -0400)]
Fix a few new (and one old) issues in the new channel_update
We need to always set lowest_inbound_channel_fees to None if there
are no channels (and we should ignore the channels which are
disabled for the purpose of lowest-fee calculations). Further, we
cannot unwrap the channel lookups as they may be for channels which
are not related to the channel we are processing an update for.
Finally, we can satisfy borrowck much easier since things are on
self instead of on a MutexGuard.
Matt Corallo [Tue, 12 May 2020 01:08:35 +0000 (21:08 -0400)]
Add assertion since we now have a pub field that is PartialEq
We previously were only able to assert that the route graph
serialization roundtrips were good by comparing the network
messages returned from them, but we can now do better as the graph
is exposed publicly via a simple datastructure that implements
PartialEq.
Matt Corallo [Tue, 12 May 2020 01:07:02 +0000 (21:07 -0400)]
Move get_addresses to network_graph and drop now-useless log_trace
Because we expose the internals we don't need a method to log
their contents anymore, and get_addresses can now avoid copying as
we expose the RwLock directly
8ce2223b6952567b28adfcc11c23c9f19c4c0324 introduced a new field in
the "Node" objects used in our functional tests - the node_id. Its
not a bad idea to cache it, but unless we want to commit and use it
everywhere, we should avoid duplicating data, especially in tests
where we'd rather exercise the underlying code than bypass it.
Matt Corallo [Sun, 3 May 2020 02:00:08 +0000 (22:00 -0400)]
Require option_static_remotekey in channel/channelmonitor.
This simplifies channelmonitor quite nicely (as expected) as we
never have to be concerned with learning data in a DataLossProtect
which is require for us to claim our funds from the latest remote
commitment transaction.
Matt Corallo [Tue, 3 Mar 2020 20:27:45 +0000 (15:27 -0500)]
Add detection of feature_static_remotekey support and print
This adds the ability to check for static_remotekey in appropriate
feature contexts and prints it at connect time. It is still
considered unknown for the purposes of requires_unknown_bits() as
we don't yet implement it.
Make channel reserve variable names less confusing.
Previous to this commit, variables such as their_channel_reserve
referred to the channel reserve that _we_ are required to keep,
(the value is initially set by the remote). Similarly,
variables such as our_channel_reserve referred to the channel
reserve that we require the remote to keep.
Change this to use local_channel_reserve / remote_channel_reserve
to refer to the the channel reserve that the local is required to keep
and the channel reserve that the remote is required to keep, respectively.
Matt Corallo [Mon, 9 Mar 2020 17:45:15 +0000 (13:45 -0400)]
Check local signtures explicitly in channel tx-generation tests
It appears the local signatures which are specified in the channel
transaction-generation tests were never checked directly (though
they were checked as a part of the overall fully-signed-transaction
tests).
Check them explicitly so that they can be updated for static remote
key.
Include tests for requires_unknown_bits and supports_unknown_bits when
an unknown even bit, odd bit, or neither is set. Refactor bit clearing
such that tests and production code share the same code path. Fix a
potential spec incompatibility (currently only exposed in testing code)
where trailing zero bytes are not removed after a bit is cleared.
Converting from InitFeatures to other Features is accomplished using
Features::with_known_relevant_init_flags. Define a more general
to_context method which converts from Features of one Context to
another.
Additionally, ensure the source context only has known flags before
selecting flags for the target context.
Refactoring the features module allowed for making code specific to
certain contexts generalizable. Specifically, KNOWN_FEATURE_MASK
is defined on Context instead of hardcoded in each method
specialization. Thus, such methods are no longer required.
Features for a given context are duplicated throughout the features
module. Use a macro for defining a Context and the applicable features
such that features only need to be defined for a Context in one place.
The Context provides bitmasks for selecting known and unknown feature
flags.
BOLT 1 and BOLT 9 refer to features as "known" if a peer understands
them. They also use the term "supported" to mean either optional or
required.
Update the features module to use similar terminology.
- Define contexts in terms of required and optional features rather than
just supported features
- Define known features as those that are optional or required
- Rename supported() constructor to known()
For completeness, clear_optional_bit for each feature is now called
clear_bits and clears both optional and required bits.
Encapsulate feature flag checking and manipulation
Each feature is represented by two bits within Features' flags field.
Working with these flags requires bitwise operations, which can be error
prone. Rather than directly checking and manipulating bits, encapsulate
the bits within each feature trait and provide mechanisms for doing so.
This removes the need to comment on which features correspond to bitwise
expressions since the expressions use feature trait identifiers instead.
With this approach, byte literals and expressions can be evaluated at
compile time still. However, for these cases, knowing which byte within
the flags that a feature corresponds to still must be determined by the
implementor.
Remove the special case where initial_routing_sync has no even bit. Now,
it (bit 2) is considered known by the implementation.
The initial_routing_sync feature is set by peer_handler whenever a full
sync of the network graph is desired. It is not explicitly set when
creating features with InitFeatures::supported().
An upcoming refactor will change supported() to known(), which will
return all features known by the implementation. Thus, the
initial_routing_sync flag will need to be set by default. This commit
makes the behavior change ahead of the refactor.
The test_upfront_shutdown_script functional test clears this feature
flag. However, the method used to clear the flag is implemented by bit
toggling. Thus, if the flag is not set the method would actually set it.
Implement the method using bit clearing instead.
Matt Corallo [Thu, 23 Apr 2020 19:43:21 +0000 (15:43 -0400)]
Don't modify LocalCommitmemntTransaction after construction
Instead of adding signatures to LocalCommitmentTransactions, we
instead leave them unsigned and use them to construct signed
Transactions when we want them. This cleans up the guts of
LocalCommitmentTransaction enough that we can, and do, expose its
state to the world, allowing external signers to have a basic
awareness of what they're signing.
Matt Corallo [Mon, 20 Apr 2020 02:59:53 +0000 (22:59 -0400)]
Batch-sign local HTLC txn with a well-doc'd API, returning sigs
1107ab06c33bd360bdee7ee64f4b690e753003f6 introduced an API to have a
ChannelKeys implementer sign HTLC transactions by calling into the
LocalCommitmentTransaction object, which would then store the tx.
This API was incredibly awkward, both because it required an
external signer trust our own internal interfaces, but also because
it didn't allow for any inspection of what was about to be signed.
Further, it signed the HTLC transactions one-by-one in a somewhat
inefficient way, and there isn't a clear way to resolve this (as
the which-HTLC parameter has to refer to something in between the
HTLC's arbitrary index, and its index in the commitment tx, which
has "holes" for the non-HTLC outputs and skips some HTLCs).
We replace it with a new function in ChannelKeys which allows us
to sign all HTLCs in a given commitment transaction (which allows
for a bit more effeciency on the signers' part, as well as
sidesteps the which-HTLC issue). This may also simplify the signer
implementation as we will always want to sign all HTLCs spending a
given commitment transaction at once anyway.
We also de-mut the LocalCommitmentTransaction passed to the
ChanKeys, instead opting to make LocalCommitmentTransaction const
and avoid storing any new HTLC-related data in it.
Matt Corallo [Sun, 19 Apr 2020 21:26:41 +0000 (17:26 -0400)]
Return Result<Signature> instead of modifying args in ChannelKeys
This cleans up sign_local_commitment somewhat by returning a
Result<Signaure, ()> over the local commitment transaction instead
of modifying the struct which was passed in.
This is the first step in making LocalCommitmentTransaction a
completely pub struct, using it just to communicate enough
information to the user to allow them to construct a signaure
instead of having it contain a bunch of logic.
This should make it much easier to implement a custom ChannelKeys
by disconnecting the local commitment transaction signing from our
own datastructures.
Matt Corallo [Sun, 19 Apr 2020 18:15:56 +0000 (14:15 -0400)]
Track signing of local txn in channelmonitor and refuse updates
In e46e183084ed858f41aa304acd78503aea1a96ed we began tracking
whether a local commitment transaction had been signed and
broadcast in OnchainTxHandler, refusing to update the local
commitment transaction state in the ChannelMonitor on that basis.
This is fine, except that it doesn't make a lot of sense to store
the full local transaction state in OnchainTxHandler - we should be
providing it the unsigned local transaction at the time we wish to
broadcast and no more (just like we do all other transaction data).
Antoine Riard [Tue, 21 Apr 2020 01:19:00 +0000 (21:19 -0400)]
Dry-up InputMaterial::Funding
As channel_value last usage was for computing feerate but as this
one is static per-commitment and will always-be following specification,
we remove it.
Antoine Riard [Fri, 10 Apr 2020 01:51:29 +0000 (21:51 -0400)]
Document exactly our CLTV sanitization policy for final incoming HTLCs
We want to avoid a third-party channel closure, where a random node
by sending us a payment expiring at current height, would trigger our
onchain logic to close the channel due to a near-expiration.
Matt Corallo [Fri, 24 Apr 2020 20:56:20 +0000 (16:56 -0400)]
Disable timeout for full_stack_target runs in CI
It seems we've recently been seeing sporadic long-running
full_stack_target cases when running honggfuzz in CI. These
shouldn't be killed (as its possible they hit an error or a deadlock,
especially since the longest-running tests probably have the most
coverage).
Matt Corallo [Tue, 14 Apr 2020 01:04:03 +0000 (21:04 -0400)]
Add test for partial-send MPP due to monitor update failure
Relatively simple test that, after a monitor update fails, we get
the right return value and continue with the bits of the MPP that
did not send after the monitor updating is restored.
Matt Corallo [Thu, 19 Mar 2020 04:34:15 +0000 (00:34 -0400)]
Add a test for timeout'ing HTLCs which claim to be a part of an MPP
This is a key test for our automatic HTLC time-out logic, as it
ensures we don't allow an HTLC which indicates we should wait for
additional HTLCs before responding to cause us to force-close a
channel due to HTLC near-timeout.
Matt Corallo [Mon, 20 Apr 2020 19:46:35 +0000 (15:46 -0400)]
Expand expect_payment_failed!() to take error codes and use it more
expect_payment_failed!() was introduced after many of the tests
which could use it were written, so we take this opportunity to
switch them over now, increasing test coverage slightly by always
checking the payment hash expected.
Matt Corallo [Thu, 9 Jan 2020 19:09:25 +0000 (14:09 -0500)]
Time out incoming HTLCs when we reach cltv_expiry (+ test)
We only do this for incoming HTLCs directly as we rely on channel
closure and HTLC-Timeout broadcast to fail any HTLCs which we
relayed onwards where our next-hop doesn't update_fail in time.
Same setup than Travis except for removing
`rm -f target/debug/lightning-*` as I do not believe
such file would exist on a fresh run.
I have not setup caching at this stage. The library is
small so I don't think it'd be that necessary/helpful.
I'd recommend to let both CI run for a bit to compare
performance and stability. The CI setup is straightforward
so I do not foresee any issue with GitHub actions.
Once happy, Travis file can be removed and branch
protection checks can be updated to block on the GitHub
actions.
You can also check the [Coverage report](https://codecov.io/gh/D4nte/rust-lightning/tree/752a58bc0441a49a0513f2cad979ad9e2621312a/lightning/src/chain) to ensure it is as expected.
Matt Corallo [Sun, 19 Apr 2020 03:13:18 +0000 (23:13 -0400)]
De-Option<> current_local_signed_commitment_tx in ChannelMonitor
Since we now are always initialised with an initial local commitment
transaction available now, we might as well take advantage of it and
stop using an Option<> where we don't need to.
Matt Corallo [Sat, 18 Apr 2020 20:35:01 +0000 (16:35 -0400)]
Delay creating outbound ChannelMonitor until funding_signed rcpt
Previously, we created the initial ChannelMonitor on outbound
channels when we generated the funding_created message. This was
somewhat unnecessary as, at that time, we hadn't yet received
clearance to broadcast our initial funding transaction, and thus
there should never be any use for a ChannelMonitor. It also
complicated ChannelMonitor a bit as, at this point, we didn't have
an initial local commitment transaction.
By moving the creation of the initial ChannelMonitor to when we
receive our counterparty's funding_signed, we can ensure that any
ChannelMonitor will always have both a latest remote commitment tx
and a latest local commitment tx for broadcast.
This also fixes a strange API where we would close a channel
unceremoniously on peer-disconnection if we hadn't yet received the
funding_signed, but we'd already have a ChannelMonitor for that
channel. While it isn't strictly a bug (some potential DoS issues
aside), it is strange that these two definitions of a channel being
open were not in sync.
Matt Corallo [Sat, 18 Apr 2020 04:10:24 +0000 (00:10 -0400)]
Add HTLC/extra data in LocalCommitmentTransaction from construction
1107ab06c33bd360bdee7ee64f4b690e753003f6 introduced some additional
metadata, including per-HTLC data in LocalCommitmentTransaction. To
keep diff reasonable it did so in ChannelMonitor after the
LocalCommitmentTransaction had been constructed and passed over the
wall, but there's little reason to do so - we should just be
constructing them with the data from the start, filled in by Channel.
This cleans up some internal interfaces a bit, slightly reduces
some data duplication and moves us one step forward to exposing
the guts of LocalCommitmentTransaction publicly in a sensible way.
Matt Corallo [Sat, 18 Apr 2020 01:26:38 +0000 (21:26 -0400)]
Drop redundant parameters in sign_local_commitment_tx
The ChanKeys is created with knowledge of the Channel's value and
funding redeemscript up-front, so we should not be providing it
when making signing requests.
Matt Corallo [Sun, 19 Apr 2020 05:16:20 +0000 (01:16 -0400)]
Avoid a few useless clone() calls in onchaintx.rs
3d640da5c343111f538f006996c13c9a98e0d9e6 looped over a new HashMap
new_claims, clone()ing entries out of it right before droppng the
whole thing. This is an obvious candidate for drain(..).
Matt Corallo [Sun, 19 Apr 2020 01:36:33 +0000 (21:36 -0400)]
Drop ChannelMonitor::write_for_watchtower
Not only was watchtower mode never implemented, but the bits that
we had were removed some time ago. It doesn't seem likely we'll
move forward with a "watchtower-mode" ChannelMonitor, instead
we'll likely have some other, separate struct for this.