To ensure a call to ConnectToPeer doesn't block the agent from
shutting down, we'll launch it in a non-waitgrouped goroutine, that
will signal when a result is returned.
Previously we would use the best known height as height hint, which
could lead to using a height hint above the actual block it confirms,
hence losing track of the confirmation.
To avoid more bugs slipping through where the logIndex is not set, we
panic to catch this. This was earlier done for Adds and the htlcCounter,
which did lead us to find the resulting retoration bug.
Earlier versions did not write the log index to disk for fee updates, so
they will be unset. To account for this we set them to to current update
log index.
We are not longer validating the max_value_in_flight field set by the
remote peer, so it is not always less than the channel capacity anymore.
We therefore make sure to cap it before advertising it.
If the max_htlc field is not found when fetching a ChannelEdgePolicy
from the DB, we treat this as an unknown policy.
This is done to ensure we won't propagate invalid data further. The data
will be overwritten with a valid one when we receive an update for this
channel.
It shouldn't be very common, but old data could be lingering in the DB
added before this field was validated.
This commit defines a new heuristic WeightedCombAttachment that takes a
set of sub-heuristics, and produces a final node score by querying the
sub-heuristics and combining the scores from them according to their
weights.
This way it will look like a regular, single heuristic to the autopilot
agemnt, but can be a more complex combination of several.
To prepare for combinning scores from multiple heuristics, we require the
scores returned from the NodeSores API to be in the range [0.0, 1.0].
The prefAttach heuristic is altered to scale the returned scores such
that the most connected node in the grpah is given a score of 1.0.
Since the chain backend and miner is now seperate nodes, the reorg test
must be slightly rewritten. We use the Btcd Node RPC to connect and
disconnect the three bitcoin nodes in question (chain backend, miner,
temp miner) to control the reorg scenario.
This tests make sure we don't reset our expected fee upate after signing
our next commitment. This test would fail without the previous set of
commits.
Instead of special casing the UpdateFee messages, we instead add them to
the update logs like any other HTLC update message. This lets us avoid
having to keep an extra set of variables to keep track of the fee
updates, and instead reuse the commit/ack logic used for other updates.
This fixes a bug where we would reset the pendingFeeUpdate variable
after signing our next commitment, which would make us calculate the new
fee incorrectly if the remote sent a commitment concurrently.
When restoring state logs, we also make sure to re-add any fee updates.
When compacting the update logs we remove any fee updates when they
remove height is passed. We do this since we'll assume fee updates are
added and removed at the same commit height, as they will apply for all
commitments following the fee update.
This commit adds conversion between the lnwire.UpdateFee message and the
new FeeUpdate PaymentDescriptor. We re-purpose the existing Amount field
in the PaymentDescriptor stuct to hold the feerate.
This commit adds a new updateType that can be used for
PaymentDescriptors: FeeUpdate. We repurpose the fields of the existing
PaymentDescriptor struct such that we can easily re-use the commit/ack
logic used for other update types also for fee updates.
Since NodeScores no longer returns fully populated AttachmentDirectives,
we make this explicit by defining a new type NodeScore that includes a
subset of what the AttachmentDirective does.
We create a new type NodeScore which is a tuple (NodeID, score). The
weightedChoice and chooseN algorithms are altered to expect this type.
This is done in order to simplify the types we are using, since we were
only using a subset of the fields in AttachmentDirective.
Since we want to combine scores from multiple heuristics, things get
complicated if the heuristics report their own individual channel sizes.
Therefore we change the NodeScores interface slightly, letting the agent
specify the wanted channel size, and let the heuristic score the nodes
accordingly.
We let the agent call ChannelBudget on its constraints directly, and
not go through the heuristic. This is needed since when we want to have
multiple active heuristics concurrently, it won't make sense anymore to
ask each of the heuristics.
The mockConstraints are also updated to act as the mockHeuristic did
before, by making it possible to control the responses it gives by
sending them on the contained channels.
To decouple the autopilot heuristic from the constraints, we start by
abstracting them behind an interface to make them easier to mock. We
also rename them HeuristicConstraints->AgentConstraints to make it clear
that they are now constraints the agent must adhere to.
This commit fixes a subtle bug within the autopilot manager, that would
cause the active pilot to not be reset in case it wasn't started
successfully.
We also make sure the associated goroutines close over the started
pilot, and not the active pilot.
This commit makes the weightedChoice algorithm take a slice of weights
instead of a map of node scores. This let us avoid costly map allocation
and iteration.
In addition we make the chooseN algorithm keep track of the remaining
nodes by keeping a slice of weights through its entire run, similarly
avoiding costly map allocation and iteration.
In total this brings the runtime of the TestChooseNSample testcase down
from ~73s to ~3.6s.
This commit adds a new service lnrpc/autopilot, that is to be used to
communicate with the running autopilot. Currently a RPC for getting the
status of the agent is included, as well as enabling/disabling at
runtime.
This commit moves the responsibility of managing the life cycle of the
autopilot from main to the autopilot Manager. It utilizes the recently
introduced autopilot Manager, and just sets up the necessary interfaces
for the Manager to properly set up the required subscriptions when
starting the agent.
This addition to the unit tests makes sure nodes that have no channels
in the graph are left out od the scored nodes, implicitly giving them a
score of 0.
Johan T. Halseth