Commit Graph

724 Commits

Author SHA1 Message Date
Olaoluwa Osuntokun
32ad632aa1
Merge pull request #956 from halseth/channel-newlocalforceclosesummary
[refactor] lnwallet/channel: add NewLocalForceCloseSummary
2018-04-09 20:47:53 -07:00
Johan T. Halseth
3d55767838
lnwallet/reservation: remove RegisterMinHTLC
We remove this method, as our minHtlc value is set using the
CommitConstraints method.
2018-04-06 10:02:19 +02:00
Johan T. Halseth
8e77e1e6eb
lnwallet/channel: add NewLocalForceCloseSummary
This commit renames ForceCloseSummary to LocalForceCloseSummary, and
adds a new method NewLocalForceCloseSummary that can be used to derive a
LocalForceCloseSummary if our commitment transaction gets confirmed
in-chain. It is meant to accompany the NewUnilateralCloseSummary method,
which is used for the same purpose in the event of a remote commitment
being seen in-chain.
2018-04-06 09:46:30 +02:00
Olaoluwa Osuntokun
065b39bc0b
Merge pull request #1025 from Roasbeef/incoming-htlc-breach-retribution
lnwallet+test: properly generate the sender HTLC script in a contract breach scenario
2018-04-05 17:58:31 -07:00
Olaoluwa Osuntokun
7da8cb2910
lnwallet: add new TestNewBreachRetributionSkipsDustHtlcs test 2018-04-05 16:20:22 -07:00
Olaoluwa Osuntokun
f79af461d3
lnwallet: in NewBreachRetribution create the htlcRetributions slice to capacity, not length
In this commit, we fix an existing bug in the NewBreachRetribution
method. Rather than creating the slice to the proper length, we instead
now create it to the proper _capacity_. As we'll now properly filter out
any dust HTLCs, before this commit, even if no HTLCs were added, then
the slice would still have a full length, meaning callers could actually
interact with _blank_ HtlcRetribution structs.

The fix is simple: create the slice with the proper capacity, and append
to the end of it.
2018-04-05 16:19:49 -07:00
Olaoluwa Osuntokun
c393475d39
lnwallet: ensure that we skip dust HTLC's in NewBreachRetribution 2018-04-04 18:41:38 -07:00
Olaoluwa Osuntokun
4e44528bff
lnwallet: fix ordering of keys when we generate the sender script during a breach
In this commit, we fix an existing within lnd. Before this commit,
within NewBreachRetribution the order of the keys when generating the
sender HTLC script was incorrect. As in this case, the remote party is
the sender, their key should be first. However, the order was swapped,
meaning that at breach time, our transaction would be rejected as it had
the incorrect witness script.

The fix is simple: swap the ordering of the keys. After this commit, the
test extension added in the prior commit now passes.
2018-04-04 18:41:37 -07:00
Olaoluwa Osuntokun
1c5d62a804
lnwallet: add new concrete error InvalidHtlcSigError for failed htlc sig validation
In this commit we add a new error: InvalidHtlcSigError. This error will
be returned when we're unable to validate an HTLC signature sent by the
remote party. This will allow other nodes to more easily debug _why_ the
signature was rejected.
2018-04-04 17:41:10 -07:00
Olaoluwa Osuntokun
b422e4ec1e
lnwallet+funding+lnd: add new config option for min accepted channel size
In this commit we add a new command line option (and a sane default) to
allow users to specify the *smallest* inbound channel that they'll
accept. Having a higher-ish limit lets users limit their channels, and
also avoid a series of very low value "spam" channels.

The new option is --minchansize, and expressed in satoshis. If we
receive an inbound channel request for a value smaller than this, then
we'll immediately reject it.
2018-04-02 16:17:58 -07:00
Olaoluwa Osuntokun
6103ccb081
lnwallet: ensure we only accept/produce sane commitments
In this commit, we add an additional check within CreateCommitTx to
ensure that we will never create or accept a commitment transaction that
wasn't valid by consensus. To enforce this check, we use the
blockchain.CheckTransactionSanity method.
2018-03-27 15:37:15 -07:00
Olaoluwa Osuntokun
af4afdb6f0
lnwallet: use btcutil.NewAmount for BTC -> SAT conversions
In this commit, we fix an existing rounding related bug in the codebase.
The RPC interface for btcd and bitcoind return values in BTC rather than
in satoshis. So in several places, we're forced to convert ourselves
manually. The existing logic attempted to do this, but didn't properly
account for rounding. As a result, our values can be off due to not
rounding incorrectly.

The fix for this is easy: simply properly use btcutil.NewAmount
everywhere which does rounding properly.

Fixes #939.
2018-03-25 19:15:18 -07:00
Johan T. Halseth
a6e7dce7b7
lnwallet/channel: reject received commitment with too many htlc sigs
This commit adds a check that will make LightningChannel reject a
received commitment if it is accompanied with too many HTLC signatures.
This enforces the requirement in BOLT-2, saying:

if num_htlcs is not equal to the number of HTLC outputs in the local commitment transaction:
    * MUST fail the channel.

A test exercising the behaviour is added.
2018-03-22 14:58:22 +01:00
Johan T. Halseth
70b86e596e
lnwallet/channel: use remote dustlimit when generating HTLC sigs
This commit fixes an issue which would arise in some cases when the
local and remote dust limits would differ, resulting in lnd not
producing the expected number of HTLC signatures. This was a result of
checking dust against the local instead of the remote dust limit.

A test exercising the scenario is added.
2018-03-22 14:58:22 +01:00
Johan T. Halseth
263d6b9c1f
lnwallet/channel: don't accept immediately on empty htlc sigs
This commit fixes an issue where we would blindly accept a commitment
which came without any accompanying HTLC signatures. A test exercising
the scenario is added.
2018-03-22 14:58:22 +01:00
Johan T. Halseth
241c79397f
lnwallet/channel: fix crash on receiving too few HTLC sigs
This commit fixes an out of bounds error that would occur in the case
where we received a new commitment where the accompanying HTLC sigs were
too few. Now we'll just reject such an commitment.

A test exercising the behavior is also added.
2018-03-22 14:58:22 +01:00
Olaoluwa Osuntokun
e0dc10b1a5
Merge pull request #824 from cfromknecht/multi-chain-coin-type
Configurable CoinType for HD Derivation
2018-03-13 18:14:30 -07:00
Olaoluwa Osuntokun
0befe41384
Merge pull request #782 from paddyquinn/master
funding: implement reservation zombie sweeper
2018-03-13 17:07:53 -07:00
Conner Fromknecht
ca7992e8dd
lnwallet/btcwallet/singer: use chainKeyScope to fetch scoped mgr 2018-03-13 16:33:46 -07:00
Conner Fromknecht
fbef815530
lnwallet/btcwallet/config: add CoinType to configuration 2018-03-13 16:33:46 -07:00
Conner Fromknecht
e760963ead
lnwallet/btcwallet/btcwallet: use coin type in key scope 2018-03-13 16:33:46 -07:00
Conner Fromknecht
e99243d86b
lnwallet/interface_test: inti btcwallet with test coin type 2018-03-13 16:33:46 -07:00
PaddyQuinn
5410725306 lnwallet: update comments 2018-03-12 19:37:18 -04:00
Conner Fromknecht
600ab9d149
lnwallet/interface_test: extend publish timeout
This commit extends the amount of time we wait
for transaction to enter the mempool from
10 to 30 seconds. The wallet's interface tests
seem to be particularly slow when run with the
race flag, a problem which is only exacerbated
by the slowness of travis.

With 10s and the race flag, I was able to repro
the issues locally fairly consistently.
2018-03-12 13:52:16 -07:00
Conner Fromknecht
5df8b52dae
multi: set initiator funding txn 2018-03-11 15:06:22 -07:00
Conner Fromknecht
2e1dcd316c
lnwallet/wallet: store funding txn during reservation 2018-03-11 13:58:41 -07:00
Olaoluwa Osuntokun
800eea931f
build+multi: switch from bolt to bbolt
In this commit, we switch from boltbd/bolt to coreos/bbolt as the
former is no longer being actively maintained.
2018-03-10 19:01:13 -08:00
Conner Fromknecht
5539e23b05
lnwallet/channel_test: convert tests to new channel API for persistence 2018-03-09 17:35:54 -08:00
Conner Fromknecht
d420266911
lnwallet/channel: expose API changes to channel state transitions 2018-03-09 17:35:53 -08:00
Conner Fromknecht
6e542d5dfa
lnwallet/channel_test: init open channels with Packager 2018-03-09 14:45:30 -08:00
Conner Fromknecht
53e4422a2e
lnwallet/channel: integrate fwdpkgs w/ in-mem buffering 2018-03-09 14:45:30 -08:00
Olaoluwa Osuntokun
b8d0df998a
lnwallet: when validating fee updates, ensure newFee < balance 2018-03-08 12:50:49 -05:00
Olaoluwa Osuntokun
ac90a8288e
lnwallet: precalculate fees in mSAT to avoid multiple conversions 2018-03-08 12:50:49 -05:00
Olaoluwa Osuntokun
217166fb10
lnwallet: within validateCommitmentSanity check for balance underflow
In this commit, we add an additional check within
validateCommitmentSanity due to the recent change to unsigned integers
for peer balances in the channel state machine. If after evaluation
(just applying HTLC updates), the balances are negative, then we’ll
return ErrBelowChanReserve.
2018-03-08 12:50:48 -05:00
Olaoluwa Osuntokun
5f5e4554cb
lnwallet: if the initiator is unable to pay fees, then consume their entire output
In this commit, we add logic to account for an edge case in the
protocol. If they initiator if unable to pay the fees for a commitment,
then their *entire* output is meant to go to fees. The recent change to
properly interpret balances as unsigned integers (within the protocol)
let to the discovery of this missed edge case.
2018-03-08 12:50:48 -05:00
Olaoluwa Osuntokun
19bc477b9a
lnwallet: update interface tests to account for recent lnwire.MilliSatoshi change
lnwire.MilliSatoshi is now a signed integer, as a result, we’ll return
a different error if our balances go to negative due to the inability
to pay a the set fee.
2018-03-08 12:50:48 -05:00
Olaoluwa Osuntokun
8425a35684
lnwallet: in NewChannelReservation ensure commit fees are payable
In this commit, we fix a bug introduced by the recent change of
lnwire.MilliSatoshi to be an unsigned integer. After this change an
integer underflow was left undetected, as a result we’ll now
momentarily cast to a signed integer in order to ensure that both sides
can pay the proper fee.
2018-03-08 12:50:47 -05:00
Olaoluwa Osuntokun
5ecef17e0f
lnwallet: modify logging to display mSAT amount if funding constrains invalid 2018-03-08 12:50:47 -05:00
Olaoluwa Osuntokun
116406c7ec
Merge pull request #769 from Roasbeef/new-lightning-key-derivation
multi: modify key derivation to be fully deterministic, remove p2pkh, wallet now witness only
2018-03-06 17:21:55 -05:00
Olaoluwa Osuntokun
18e9475a9a
lnwallet/btcwallet: grab best header timestamp directly from wallet
In this commit, we modify the way we obtain the current best header
timestamp. In doing this, we fix an intermittent flake that would pop
up at times on the integration tests. This could occur as if the wallet
was lagging behind the chain backend for a re-org, then a hash that the
backend knew of, may not be known by the wallet.

To remedy this, we’ll take advantage of a recent change to btcwallet to
actually include the timestamp in its sync state.
2018-03-06 16:04:07 -05:00
Olaoluwa Osuntokun
d98e7a730d
lnwallet: update integration tests due to recent API changes 2018-03-06 16:04:06 -05:00
Olaoluwa Osuntokun
a144018e98
lnwallet: remove p2pkh as an address type, wallet is now pure segwit
In this commit, we modify the mechanics of the wallet to only allow
derivation of segwit-like addresses. Additionally, the ConfirmedBalance
method on the WalletController now only has a single argument, as it’s
assumed that the wallet is itself only concerned with segwit outputs.
2018-03-06 16:04:03 -05:00
Olaoluwa Osuntokun
0d1a40fb46
lnwallet: update tests due to recent API changes 2018-03-06 16:04:03 -05:00
Olaoluwa Osuntokun
5b063a0691
lnwallet: modify the way we derive revocation roots to be deterministic
In this commit, we modify the way we generate the secrets for
revocation roots to be fully deterministic. Rather than use a special
key and derive all sub-roots from that (mixing in some “salts”), we’ll
use the proper keychain.KeyFamily instead. This ensures that given a
static description of the channel, we’re able to re-derive our
revocation root properly.
2018-03-06 16:04:02 -05:00
Olaoluwa Osuntokun
a41f00e2d6
lnwallet: update funding flow to utilize keychain.KeyRing
In this commit, we modify the funding flow process to obtain all keys
necessary from the keychain.KeyRing interface. This ensure that all
keys we generate are fully deterministic.
2018-03-06 16:04:02 -05:00
Olaoluwa Osuntokun
fe12c908f8
lnwallet: add an instance of keychain.SecretKeyRing to the Config struct 2018-03-06 16:04:02 -05:00
Olaoluwa Osuntokun
22ee0a7576
lnwallet: update state machine due to channeldb and SignDescriptor changes 2018-03-06 16:04:02 -05:00
Olaoluwa Osuntokun
af1fd65655
lnwallet/btcwallet: update Signer implementation to use keychain.KeyDescriptor 2018-03-06 16:04:01 -05:00
Olaoluwa Osuntokun
a0dc06b954
lnwallet: update script_utils and tests due to SignDescriptor API change 2018-03-06 16:04:01 -05:00
Olaoluwa Osuntokun
73dd1536ed
lnwallet: update signdescriptor_test.go due to latest API changes 2018-03-06 16:04:00 -05:00