In preparation for adding the new ChannelAnnouncement2 message along
with a ChannelAnnouncement interface, we rename the existing message to
ChannelAnnouncement1.
In preparation for adding a new message, AnnounceSignatures2 along with
an AnnounceSignatures interface, we rename the existing message to
AnnounceSignatures1.
This commit hooks up the banman to the gossiper:
- peers that are banned and don't have a channel with us will get
disconnected until they are unbanned.
- peers that are banned and have a channel with us won't get
disconnected, but we will ignore their channel announcements until
they are no longer banned. Note that this only disables gossip of
announcements to us and still allows us to open channels to them.
This commit is a large refactor that moves over various responsibilities
from the ChannelRouter to the graph.Builder. These include all graph
related tasks such as:
- graph pruning
- validation of new network updates & persisting new updates
- notifying topology update clients of any changes.
This is a large commit but:
- many of the files are purely moved from `routing` to `graph`
- the business logic put in the graph Builder is copied exactly as is
from the ChannelRouter with one exception:
- The ChannelRouter just needs to be able to call the Builder's
`ApplyChannelUpdate` method. So this is now exported and provided to
the ChannelRouter as a config option.
- The trickiest part was just moving over the test code since quite a
bit had to be duplicated.
In preparation for a more complex function signature for set node
announcement, separate get and set so that readonly callers don't need
to handle the extra arguments.
This commit adds a simple struct `futureMsgCache` that embeds a lru
cache with the message ID. A unit test is added to check the eviction
behaves as expected.
We require channel updates to have the max HTLC message flag set.
Several flows need to pass that check before channel updates are
forwarded to peers:
* after channel funding: `addToRouterGraph`
* after receiving channel updates from a peer:
`ProcessRemoteAnnouncement`
* after we update channel policies: `PropagateChanPolicyUpdate`
We rename `ChanUpdateOptionMaxHtlc` to `ChanUpdateRequiredMaxHtlc`
as with the latest changes it is now required.
Similarly, rename `validateOptionalFields` to
`ValidateChannelUpdateFields`, export it to use it in a later commit.
This commit changes the sending of anns from using separate goroutines
to always sending both local and remote announcements in the same
goroutine. In addition, the local announcements are always sent first.
This change is to fix the following case:
1. Alice and Bob have a channel
2. Alice receives Bob's NodeAnnouncement
3. Alice goes to broadcast the channel
4. The broadcast is split into a local and remote broadcast due to PR
#7239. Bob's NodeAnnouncement is in the remote batch. Everything else
(ChannelAnnouncement, ChannelUpdate x2, and Alice's NodeAnnouncement)
is in the local batch.
5. The remote batch (containing Bob's NodeAnnouncement) runs before the
local batch since they are spawned in separate goroutines. This means
that Alice sends Carol the NodeAnnouncement before Carol knows of the
channel.
In step 2), Bob's NodeAnnouncement (isRemote = true) replaces Bob's
NodeAnnouncement that Alice was going to relay (isRemote = false) after
processing the AnnouncementSignatures.
This commit refactors the method `sendBatch` into `sendLocalBatch` and
`sendRemoteBatch` for clarity. The batch size calculation is also moved
into `splitAnnouncementBatches`.
In this commit, we modify our gossip broadcast logic to ensure that we
always will send out our own gossip messages regardless of the
filtering/feature policies of the peer.
Before this commit, it was possible that when we went to broadcast an
announcement, none of our peers actually had us as a syncer peer (lnd
terminology). In this case, the FilterGossipMsg function wouldn't do
anything, as they don't have an active timestamp filter set. When we go
to them merge the syncer map, we'd add all these peers we didn't send
to, meaning we would skip them when it came to broadcast time.
In this commit, we now split things into two phases: we'll broadcast
_our_ own announcements to all our peers, but then do the normal
filtering and chunking for the announcements we got from a remote peer.
Fixes https://github.com/lightningnetwork/lnd/issues/6531
Fixes https://github.com/lightningnetwork/lnd/issues/7223
Fixes https://github.com/lightningnetwork/lnd/issues/7073
In this commit, we add a new option for the existing confirmation
notification system that optionally allows the caller to specify that a
block should be included as well.
The only quirk w/ the implementation here is the neutrino backend:
usually we get filtered blocks, we so need to first fetch the block
again so we can deliver the full block to the notifier. On the notifier
end, it'll only be checking for the transactions we care about, to
sending a full block doesn't affect the correctness.
We also extend the `testBatchConfirmationNotification` test to assert
that a block is only included if the caller specifies it.
An OptionalMsgField has been added that allows outside subsystems
to provide a short channel id we should insert into a ChannelUpdate
that we then sign and send to our peer.
When the gossiper receives a ChannelUpdate, it will query the
alias manager by the passed-in FindBaseByAlias function to determine
if the short channel id in the ChannelUpdate points to a known
channel. If this lookup returns an error, we'll fallback to using
the original id in the ChannelUpdate when querying the router.
The lookup and potential fallback must occur in order to properly
lock the multimutex, query the correct router channels, and rate
limit the correct short channel id. An unfortunate side effect of
receiving ChannelUpdates from our peer that reference on of our
aliases rather than the real SCID is that we must store this policy.
Yet it is not broadcast-able. Care has been taken to ensure the
gossiper does not broadcast *any* ChannelUpdate with an alias SCID.
The cachedNetworkMsg uses the new processedNetworkMsg struct. This
is necessary so that delete-and-reinsert in the funding manager
doesn't process a ChannelUpdate twice and end up in a deadlock since
the err chan is no longer being used.
This commit was previously split into the following parts to ease
review:
- 2d746f68: replace imports
- 4008f0fd: use ecdsa.Signature
- 849e33d1: remove btcec.S256()
- b8f6ebbd: use v2 library correctly
- fa80bca9: bump go modules
To simplify the message signing API even further, we refactor the
lnwallet.MessageSigner interface to use a key locator instead of the
public key to identify which key should be signed with.
Adds an optional tx parameter to ForAllOutgoingChannels and FetchChannel
so that data can be queried within the context of an existing database
transaction.
In this commit, we update the existing zombie resurrection test to
ensure that if we prune an edge and another pubkey is marked as nil,
that we only accept a resurrection channel update from the node the we
originally pruned if the pruning decision was one sided.
The recently added gossip throttling was shown to be too aggressive,
especially with our auto channel enable/disable signaling. We switch to
a token bucket based system instead as it's based on time, rather than a
block height which isn't constantly updated at a given rate.
AFAICT it's not possible to flip back from bein synced_to_chain, so we
remove the underlying call that could reflect this. The method is moved
into the test file since it's still used to test correctness of other
portions of the flow.
As similarly done with premature channel announcements, we'll no longer
allow premature channel updates to be rebroadcast once mature. This is
no longer necessary as channel announcements that we're not aware of are
usually broadcast to us with their accompanying channel updates.
In this commit, we add a new option to toggle gossip rate limiting. This
new option can be useful in contexts that require near instant
propagation of gossip messages like integration tests.
This change was largely motivated by an increase in high disk usage as a
result of channel update spam. With an in memory graph, this would've
gone mostly undetected except for the increased bandwidth usage, which
this doesn't aim to solve yet. To minimize the effects to disks, we
begin to rate limit channel updates in two ways. Keep alive updates,
those which only increase their timestamps to signal liveliness, are now
limited to one per lnd's rebroadcast interval (current default of 24H).
Non keep alive updates are now limited to one per block per direction.
This commit adds a reset() closure to the kvdb.View function which will
be called before each retry (including the first) of the view
transaction. The reset() closure can be used to reset external state
(eg slices or maps) where the view closure puts intermediate results.
This commit moves all localized instances of mock implementations of
the Signer interface to the lntest/mock package. This allows us to
remove a lot of code and have it housed under a single interface in
many cases.
