Jeffrey Czyz [Fri, 2 Aug 2024 14:39:52 +0000 (09:39 -0500)]
Don't include HMAC in Refund paths
Refunds are typically communicated via QR code, where a smaller size is
desirable. Make the HMAC in OutboundPayment data optional such that it
is elided from blinded paths used in refunds. This prevents abandoning
refunds if the reader sends an invoice_error instead of an invoice
message. However, this use case isn't necessary as the corresponding
outbound payment will either timeout when the refund expires or can be
explicitly abandoned by the creator.
A BOLT12 payment may be abandoned when handling the invoice or when
receiving an InvoiceError message. When abandoning the payment, don't
use UserAbandoned as the reason since that is meant for when the user
calls ChannelManager::abandon_payment.
When making an outbound BOLT12 payment, multiple invoices may be
received for the same payment id. Instead of abandoning the payment when
a duplicate invoice received, simply ignore it without responding with
an InvoiceError. This prevents abandoning in-progress payments and
sending unnecessary onion messages.
Before abandoning a payment when receiving an InvoiceError, verify that
the PaymentId included in the OffersContext with the included HMAC. This
prevents a malicious actor sending an InvoiceError with a known payment
id from abandoning our payment.
When receiving an InvoiceError in response to an InvoiceRequest, the
corresponding payment should be abandoned. Add an HMAC to
OffersContext::OutboundPayment such that the payment ID can be
authenticated prior to abandoning the payment.
An HMAC needs to be included in OffersContext::OutboundPayment to
authenticate the included PaymentId. Implement Readable and Writeable to
allow for this.
When receiving an InvoiceError in response to an InvoiceRequest, the
corresponding payment should be abandoned. Add functions for
constructing and verifying an HMAC over a Payment ID to allow for this.
Matt Corallo [Fri, 9 Aug 2024 15:27:38 +0000 (15:27 +0000)]
`rustfmt` new files added in the past few commits
The past handful of commits were mostly moving code around, so to
aid reviewers violated our `rustfmt` rules. Here we rectify that by
`rustfmt`'ing the newly-added files.
Matt Corallo [Fri, 9 Aug 2024 15:42:48 +0000 (15:42 +0000)]
Prepare to `rustfmt` newly added files
In the next commit we'll `rustfmt` newly-added files, but before
we do so we clean up some code so that the resulting files won't be
quite as absurd. We also exclude the new `invoice_utils.rs` file,
as it needs quite substantial cleanups.
Matt Corallo [Fri, 9 Aug 2024 02:45:55 +0000 (02:45 +0000)]
Provide the signer with a full `RawBolt11Invoice` to sign
Now that the `lightning` crate depends on the `lightning-invoice`
crate, there's no reason to have the `sign_invoice` method take raw
base32 field elements as we can now give it a real
`RawBolt11Invoice`, which we do here.
This simplifies the interface and avoids a
serialization-deserialization roundtrip when signing invoices in a
validating signer.
Matt Corallo [Fri, 9 Aug 2024 01:29:48 +0000 (01:29 +0000)]
Swap the dep order between `lightning` and `lightning-invoice`
`lightning-invoice` previously had a dependency on the entire
`lightning` crate just because it wants to use some of the useful
types from it. This is obviously backwards and leads to some
awkwardness like the BOLT 11 invoice signing API in the `lightning`
crate taking a `[u5]` rather than a `Bolt11Invoice`.
Here we finally rectify this issue, swapping the dependency order
and making `lightning` depend on `lightning-invoice` rather than
the other way around.
This moves various utilities which were in `lightning-invoice` but
relied on `lightning` payment types to make payments to where they
belong (the `lightning` crate), but doesn't bother with integrating
them well in their new home.
Matt Corallo [Fri, 9 Aug 2024 01:13:25 +0000 (01:13 +0000)]
Add a `lightning-types` dependency to `lightning-invoice`
`lightning-invoice` currently has a dependency on the entire
`lightning` crate just because it wants to use some of the useful
types from it. This is obviously backwards and leads to some
awkwardness like the BOLT 11 invoice signing API in the `lightning`
crate taking a `[u5]` rather than a `Bolt11Invoice`.
This takes tees us up for the final step, adding a
`lightning-types` dependency to `lightning-invoice` and using it
for imports rather than the `lightning` crate.
Matt Corallo [Fri, 9 Aug 2024 13:20:35 +0000 (13:20 +0000)]
Use `check_added_monitors` test utility in invoice utils tests
In a coming commit, the `lightning-invoice::utils` module will move
to the `lightning` crate, causing its tests to be included in the
global lockorder tests done in that crate. This should be fine,
except that the `lightning-invoice::utils` module currently holds
the `added_monitors` lock too long causing lockorder violations.
Instead, this commit replaces the legacy monitors-added test with
the `check_added_monitors` test utility.
Matt Corallo [Fri, 9 Aug 2024 01:26:21 +0000 (01:26 +0000)]
Move `UntrustedString` and `PrintableString` to `lightning-types`
`lightning-invoice` currently has a dependency on the entire
`lightning` crate just because it wants to use some of the useful
types from it. This is obviously backwards and leads to some
awkwardness like the BOLT 11 invoice signing API in the `lightning`
crate taking a `[u5]` rather than a `Bolt11Invoice`.
This takes one more step, moving the `UntrustedString` and
`PrintableString` types to `lightning-types`.
Matt Corallo [Fri, 9 Aug 2024 14:20:50 +0000 (14:20 +0000)]
Move `Features` into `lightning-types`
`lightning-invoice` currently has a dependency on the entire
`lightning` crate just because it wants to use some of the useful
types from it. This is obviously backwards and leads to some
awkwardness like the BOLT 11 invoice signing API in the `lightning`
crate taking a `[u5]` rather than a `Bolt11Invoice`.
This takes one more step, moving the `Features` types from
`lightning` to `lightning-types`.
Matt Corallo [Fri, 9 Aug 2024 14:18:09 +0000 (14:18 +0000)]
Replace usages of `Features::is_subset` and remove it
It turns out all the places we use `Features::is_subset` we could
as well be using `Features::requires_unknown_bits_from`. Further,
in the next commit `Features` will move to a different crate so any
methods which the `lightning` crate uses will need to be public. As
the `is_subset` API is prety confusing (it doesn't consider
optional/required bits, only whether the bits themselves are
strictly a subset) it'd be nice to not have to expose it, which is
enabled here.
Matt Corallo [Fri, 9 Aug 2024 00:29:25 +0000 (00:29 +0000)]
Move `Rout{ingFees,eHint{,Hop}}` to `lightning-types`
`lightning-invoice` currently has a dependency on the entire
`lightning` crate just because it wants to use some of the useful
types from it. This is obviously backwards and leads to some
awkwardness like the BOLT 11 invoice signing API in the `lightning`
crate taking a `[u5]` rather than a `Bolt11Invoice`.
This takes one more step, moving the routing types
`lightning-invoice` uses into `lightning-types`.
Matt Corallo [Fri, 9 Aug 2024 00:13:41 +0000 (00:13 +0000)]
Move `Payment{Hash,Preimage,Secret}` into a new crate
`lightning-invoice` currently has a dependency on the entire
`lightning` crate just because it wants to use some of the useful
types from it. This is obviously backwards and leads to some
awkwardness like the BOLT 11 invoice signing API in the `lightning`
crate taking a `[u5]` rather than a `Bolt11Invoice`.
This is the first step towards fixing that - moving the common
types we need into a new `lightning-types` crate which both can
depend on.
Since we're using a new crate and can't depend on the existing
`lightning` hex utility to implement `Display`, we also take this
opportunity to switch to the new `Display` impl macro in
`hex_conservative`.
Matt Corallo [Tue, 14 Nov 2023 22:19:19 +0000 (22:19 +0000)]
Include rounded msat balances in `Balance::ClaimableOnChannelClose`
If we're gonna push users towards using `Balance` to determine
their current balances, we really need to provide more information,
including msat balances.
Here we add rounded-out msat balances to the pre-close balance
information
Matt Corallo [Fri, 29 Sep 2023 17:54:24 +0000 (17:54 +0000)]
Add tx fee information to `Balance::ClaimableOnChannelClose`
`Balance::ClaimableOnChannelClose` excludes the commitment
transaction fee, which makes it hard to use for current balance
calculation. Here we add it, setting the value to zero for inbound
channels (i.e. ones for which we don't pay the fee).
Matt Corallo [Thu, 28 Sep 2023 23:02:26 +0000 (23:02 +0000)]
Include an `outbound_payment` flag in `MaybeTimeoutClaimableHTLC`
When the user is fetching their current balances after forwarding a
payment (before it clears), they'll see a
`MaybePreimageClaimableHTLC` and a `MaybeTimeoutClaimableHTLC` but
if they sum up their balance using
`Balance::claimable_amount_satoshis` neither will be included.
Obviously, exactly one of the two balances should be included - one
of the two resolutions should happen in our favor. This causes our
visible balance to fluctuate up and down by the full value of any
HTLCs we're in the middle of forwarding, which is incredibly
confusing to see. If we want to stop the fluctuations, we need to
pick one of the two balances to include. The obvious candidate is
`MaybeTimeoutClaimableHTLC` as it is the lower of the two, and
represents our balance without the fee we'd receive from the
forward.
Sadly, if we always include it, we'll end up also including any
HTLCs which we've sent but which haven't yet been claimed by their
recipient, which is the wrong behavior.
Luckily, we have access to the `Option<HTLCSource>` while walking
HTLCs, which allows us to add an `outbound_payment` flag to
`MaybeTimeoutClaimableHTLC`. This allows us to only include
forwarded payments in `claimable_amount_satoshis`.
Sadly, even with this in place our balance still fluctuates by the
changes in the commitment transaction fees we have to pay during
forwarding, but addressing that is left for later.
Matt Corallo [Mon, 12 Aug 2024 01:03:00 +0000 (01:03 +0000)]
Stop relying on a `Clone`able `NetworkGraph` ref in `DefaultRouter`
While there's not really much harm in requiring a `Clone`able
reference (they almost always are), it does make our bindings
struggle a bit as they don't support multi-trait bounds (as it
would require synthesizing a new C trait, which the bindings don't
do automatically). Luckily, there's really no reason for it, and we
can just call the `DefaultMessageRouter` directly when we want to
route a message.
We've carried this patch for a while on the bindings branch, but
there's not a strong reason it can't go upstream.
Use the `hex-conservative` crate directly from `bitcoin` instead of from
`hashes`. Although it makes no real difference it is slightly more clear
and more terse.
The `hex` crate is re-exported by `rust-bitcoin` so we can get it from
there instead of explicitly depending on it. Doing so reduces the
maintenance burden and helps reduce the likelyhood of getting two
versions in the dependency graph.
Duncan Dean [Fri, 2 Aug 2024 11:44:28 +0000 (13:44 +0200)]
Set default ChannelHandshakeLimits::min_funding_satoshis to 1000
The original default value of 0 was inconsistent with the minimum requirement
of 1000 satoshis in ChannelHandshakeConfig::their_channel_reserve_proportional_millionths.
Elias Rohrer [Thu, 1 Aug 2024 16:30:41 +0000 (11:30 -0500)]
Protect against Core's Merkle leaf node weakness
Bitcoin Core's Merkle tree implementation has no way to discern between
internal and leaf node entries. As a consequence it is susceptible to an
attacker injecting additional transactions by crafting 64-byte
transactions matching an inner Merkle node's hash (see
https://web.archive.org/web/20240329003521/https://bitslog.com/2018/06/09/leaf-node-weakness-in-bitcoin-merkle-tree-design/).
To protect against this (highly unlikely attack vector), we check that
the transaction isn't 64 bytes in length, and skip it otherwise.
cooltexture [Wed, 7 Aug 2024 21:54:11 +0000 (23:54 +0200)]
Fix #3209
There was a issue with the ci/check-compiles.sh.
It would return a warning due to links not being enclosed in <>.
Fixed the issue by enclosing the links.
The `rust-bitcoin` project is working towards making the public API
separate from the directory structure; eventually the
`bitcoin::blockdata` will go away, to make maintenance easier here stop
using the `blockdata` module.
Do not run the formatter, so as to make review easier. This patch was
created mechanically using:
search-and-replace bitcoin::blockdata bitcoin
and having defined
```bash
search-and-replace () {
if (($# != 2))
then
echo "Usage: $0 <this> <that>"
return
fi
local this="$1"
local that="$2"
for file in $(git grep -l "$this")
do
perl -pi -e "s/$this/$that/g" "$file"
done
}
```
Matt Corallo [Wed, 7 Aug 2024 14:31:01 +0000 (14:31 +0000)]
Minor doc tweaks to `MonitorHolder`
036c31c9d0b6a243fa33aa5f8d5148d2ca065617 introduced some minor doc
changes but failed to imrpove the docs that is was changing fully
which this does. Suggested by @tnull in review.
Matt Corallo [Mon, 5 Aug 2024 16:59:57 +0000 (16:59 +0000)]
Add a script to automatically `rustfmt` all required files
As we now require `rustfmt` pass on a subset of our files, its
helpful to have a script which will automatically format any
required files so that contributors don't need to think too hard
about it.
Matt Corallo [Sun, 4 Aug 2024 15:15:11 +0000 (15:15 +0000)]
Make `send_payment_with_route` take `Route` by value
Now that `ChannelManager::send_payment_with_route` is deprecated,
we don't care too much about making it as effecient as possible, so
there's not much cost to making it take `Route` by value. This
avoids bindings being unsure if the by-reference `Route` passed
needs to outlive the `ChannelManager` itself or if it only needs to
outlive the method call, creating some call overhead by forcing a
`Route::clone`, but avoiding a memory leak.
Matt Corallo [Sun, 4 Aug 2024 15:06:12 +0000 (15:06 +0000)]
Mark `ChannelManager::send_payment_with_route` as deprecated
We probably should have done this long ago a release or two after
adding `send_payment`, but we didn't and the second best time is
now.
`send_payment_with_route` has particularly hard to use retry
semantics that make it unsuitable for real use. Once we get the
last of our users off of it, we'll want to remove it (or at least
mark it test-only), but we should start by deprecating it.
Elias Rohrer [Thu, 1 Aug 2024 16:51:53 +0000 (11:51 -0500)]
Pin `tokio` in CI to fix MSRV
The recently released `tokio` 1.39 bumped their MSRV to rustc 1.70.
Here, we pin the `tokio` version to 1.38 for users that require to
maintain our MSRV of rustc 1.63.
Now that ChannelManager uses a known OffersContext when creating blinded
paths, OffersContext::Unknown is no longer needed. Remove it and update
OffersMessageHandler to us an Option, which is more idiomatic for
signifying whether a message was delivered with or without an
OffersContext.
Instead of using OffersContext::Unknown for the Bolt12Invoice reply path
use OffersContext::InboundPayment to include the payment hash.
OffersContext::Unknown will be removed in another commit.
When creating a Bolt12Invoice in fuzz tests, use
BlindedPath::new_for_payment instead of BlindedPath::new_for_message.
This way PaymentContext is used instead of MessageContext, as is more
realistic though should not affect the test. This allows us to remove
OffersContext::Unknown.
Best practice is to use different IV bytes for different contexts.
Update Offer and Refund metadata computation to use different IV bytes
when the metadata is included in a blinded path. For invoice requests,
the metatdata will always be in the blinded path, so it remains the
same.
In an upcoming commit, the iv_bytes used in MetadataMaterial will vary
depending on when whether a blinded path is included in the
corresponding message. Delay adding into MetadataMaterial::hmac as
otherwise the HmacEngine would need to be re-initialized using an
ExpandedKey, which won't be readily available.
Matt Corallo [Mon, 22 Jul 2024 20:03:41 +0000 (20:03 +0000)]
Enforce segwit inputs for all "safe" funding transactions
8403755a2a524beb9f6c8951f51dd60f7c54c912 introduced a separate path
for funding a channel without a full funding transaction, relying
on users to manually broadcast the funding tx. One of the major
things that makes this path less safe is that for other paths we're
supposed to validate that all inputs have witnesses, making the
funding transaction (likely) txid-non-malleable.
However, in one of several rewrites of that commit the funding tx
tests ended up getting elided in some call paths, which is fixed
here.
InvoiceRequest and Refund have payer_metadata which consists of an
encrypted payment id and a nonce used to derive its signing keys and
authenticate any corresponding invoices. Now that the blinded paths
include this data in their OffersContext, remove the nonce as it is now
redundant. Keep the encrypted payment id as some data is needed in the
payer metadata according to the spec. This saves space and prevents
de-anonymization attacks as along as the nonce isn't revealed.
When authenticating that an invoice is for a valid invoice request, the
payer metadata is needed. Some of this data will be removed in the next
commit and instead be included in the message context of the request's
reply path. Add this data to Event::InvoiceReceived so that asynchronous
invoice handling can verify properly.
When a Bolt12Invoice is handled with an OfferContext, use both the
containing payment_id and nonce to verify that it is for a pending
outbound payment. Previously, the nonce the payment_id were taken from
the payer_metadata and the latter was compared against the payment_id
in the OfferContext. The payer_metadata thus no longer needs to include
either when a blinded path is used. However, some payer_metadata will
still be needed as per the spec.
To authenticate that a Bolt12Invoice is for a valid InvoiceRequest or
Refund, include the nonce from the payer_metadata in the InvoiceRequest
reply path or Refund::paths, respectively. This can be used to prevent
de-anonymization attacks where an attacker sends invoices using
self-constructed paths to nodes near the blinded paths' introduction
nodes.
Invoices are authenticated by checking the payer metadata in the
corresponding invoice request or refund. For all invoices requests and
for refunds using blinded paths, this will be the encrypted payment id
and a 128-bit nonce. Allows checking the unencrypted payment id and
nonce explicitly instead of the payer metadata. This will be used by an
upcoming change that includes the payment id and nonce in the invoice
request's reply path and the refund's blinded paths instead of
completely in the payer metadata, which mitigates de-anonymization
attacks.
When using RefundBuilder::deriving_payer_id, the nonce generated needs
to be the same one included in any RefundBuilder::paths. This is because
the nonce is used along with the refund TLVs to derive a payer id and
will soon be used to authenticate any invoices.
When using InvoiceRequestBuilder::deriving_payer_id, the nonce generated
needs to be the same one included in any reply path. This is because the
nonce is used along with the invoice request TLVs to derive a payer id.
While this data is also included in the payer_metadata, including it in
the blinded path would allow reducing the amount of data needed there to
just enough to provide entropy (i.e., 16 bytes). This is more important
for Refund because it can be transmitted via a QR code. But using the
same payer_metadata structure for both InvoiceRequest and Refund would
be beneficial to avoid more code.