In this commit, we add a new integration tests to exercise the fix
introduced in the prior commit. In this test, we reconstruct a scenario
for a 3rd party to sweep an anchor spend after force closing, causing a
prior spend we had to be invalidated. Without the prior commit, this test
fails as the original anchor sweep is still found in the wallet.
Before this commit, we we were trying to sweep an anchor output, and
that output was spent by someone else (not the sweeper), then we would
report this back to the original resolver (allowing it to be cleaned
up), and also remove the set of inputs spent by that transaction from
the set we need to sweep.
However, it's possible that if a user is spending unconfirmed outputs,
then the wallet is holding onto an invalid transaction, as the outputs
that were used as inputs have been double spent elsewhere.
In this commit, we fix this issue by recursively removing all descendant
transactions of our past sweeps that have an intersecting input set as
the spending transaction. In cases where a user spent an unconfirmed
output to funding a channel, and that output was a descendant of the now
swept anchor output, the funds will now properly be marked as available.
Fixes#6241
In this commit, we add a new field to the WalletBalance call that
permits users to account for the set of outputs that may be locked due
to a pending transaction. Without this field any time users locked
outputs for things like PSBT signing, then they disappear from the
WalletBalance call, which may cause a panic.
In this commit we update go-pretty to use the latest v6 version of the
library. The existing version we used had a reported vulnerability.
Updating to this newest version also helped to shrink our set of
indirect dependencies.
In the future we can use
https://github.com/lightningnetwork/lnd/pull/5870 to detect/flag these
issues automatically.
Fixes#6293
In this commit, we increase the legacy fee limit threshold (the amount
below which we'll allow 100% of funds to go to fees for the non-v2 RPC
calls) from 50 sats to 1k sats.
In this commit, we expose a new monitoring option to allow users to
export gRPC performance metrics. These metrics can be used to see how
long certain calls are taking, the total amount of time spent handling
calls, broken down by service and also call.
This option consumes additional memory and disk space for the Prometheus
server, which is why we're opting to make it an optional flag.
For older nodes, this bucket was never created, so we'll get an error if
we try and query it. In this commit, we catch this error like we do when
a given channel doesn't have the information (but the bucket actually
exists).
Fixes#6155
This was not properly enforced and would be a spec violation on the
peer's end. Also re-use a pong buffer to save on heap allocations if
there are a lot of peers. The pong buffer is only read from, so this
is concurrent safe.
In this commit, we fix an inadvertent memory leak by ensuring we always
use `defer` to clean up the allocated objects/memory we use to be
notified of new blocks to update what we send within the set of ping
headers.
A further optimization here would be using a single global block epoch
housed within the server, that all peer `pingHandler` goroutines use
directly.
Fixes#6143.
In this commit, we switch to always sending a channel type when we're in
explicit mode. This is compatible with prior versions of lnd as they
won't send a channel type, and we'll just arrive at the same type via
the existing implicit funding.
Fixes https://github.com/lightningnetwork/lnd/issues/6067
In this commit, we fix an API inconsistency introduced by a recent fix.
Without this commit, it would be possible for a local user to attempt an
explicit funding type, but then _fallback_ to implicit negotiation if
the remote party didn't have the bit set.
We resolve this by adding a new bit of information to ensure that if
we're the funder and want explicit chan negotiation, we don't settle
for anything less. A new test case has been added to exercise this
behavior.
To recap for posterity, the following issues were found and fixed in the
negotiation logic:
1. If the remote party sent a channel type (in accept channel), but we
didn't, then we would error out. We no longer error out and instead
ensure the channel type they sent matches what we derived via
implicit funding.
2. If the remote party _did not_ have the bit set, but they sent a
chan type (in open channel), we would error out. We no longer error
out, but instead will fall back to implicit negotiation.
Ultimately we want to eventually flip the explicit funding bit to
_required_ to eliminate any other future ambiguities and also ensure
that it isn't possible to inadvertently fall back to implicit funding,
when the _user_ expects explicit funding.
In this commit, we fix a bug that would cause newly updated nodes to be
unable to start up, if they have an older channel that was closed before
we started to store all the historical state for each channel.
The issue is that we started to write the complete state to disk, but
newer channels don't have it, so when we try to supplement the
resolvers, we run into this error.
Ultimately, we only need this new supplemented information for script
enforcement channels. Ideally we would instead check the channel type
there instead, but it doesn't appear to be available in this context as
is, without further changes.
Fixes https://github.com/lightningnetwork/lnd/issues/6001.
This fixes an issue where if one tries to unset a feature like anchors,
and other feature depend on it, then `lnd` fails to start as it realizes
that its dependnacy set is inconsistent.
Fixes https://github.com/lightningnetwork/lnd/issues/6002
In this commit, we remove the `--amp-reuse` flag as with the latest
version of lnd that implements properly re-useable AMP invoices, this
flag is no longer needed.
Unfortunately, we can't use the graph cache directly here as it doesn't
include all the information we need, since it only includes the minimal
amount of information needed for path finding.
Instead, we use a simple mutex guarded cache that evicts itself after a
certain interval. The default is set small enough that most users
shouldn't really notice. We also provide a way to disable the cache, and
also disable the cache in the itests.
Fixes https://github.com/lightningnetwork/lnd/issues/1232
This commit, adds a new ForEachNode method to the channel graph cache
that assumes the contents won't be modified. This is generally useful,
and will be used in a later commit to optimize some heavy RPC calls.
Turns out we need it right now to handle some low latency race
conditions in our integration tests, so we'll opt to simply cap the size
of it to a low amount. We use a basic LRU caching mechainsm.
Fixes https://github.com/lightningnetwork/lnd/issues/5076
Olaoluwa Osuntokun