Commit Graph

530 Commits

Author SHA1 Message Date
nsa
3dde7bc4e2 routing: adding the BlockPadding value to sendpayment
This commit adds the BlockPadding value (currently 3) to sendpayment
calls so that if some blocks are mined while the htlc is in-flight, the
exit hop won't reject it.
2019-07-23 21:03:11 -04:00
Olaoluwa Osuntokun
8c389d13f9
Merge pull request #3197 from breez/optimize_prune_zombie_channels
Optimize prune zombie channels
2019-07-18 20:56:56 -07:00
Roei Erez
da9edc876a router: only prune disabled channels when AssumeChannelValid=true. 2019-07-18 17:05:20 +03:00
Joost Jager
34b264a3d8
routing: create PathFindingConfig for global parameters 2019-07-18 15:49:25 +02:00
Olaoluwa Osuntokun
92e14af72a
Merge pull request #3298 from joostjager/globalize-mc
routing: globalize mc
2019-07-16 17:39:36 -07:00
Roei Erez
9781ea0082 router: Optimize pruneZombieChannels.
The current approach iterates all channels in the graph in order to
filter those in need. This approach is time consuming, several seconds
on my mobile device for ~40,000 channels, while during this time the
db is locked in a transaction.

The proposed change is to use an existing functionality that utilize the
fact that channel update are saved indexed by date. This method enables
us to go over only a small subset of the channels, only those that
were updated before the "channel expiry" time and further filter
them for our need.
The same graph that took several seconds to prune was pruned, after
the change, in several milliseconds.

In addition for testing purposes I added Initiator field to the
testChannel structure to reflect the channeldEdgePolicy direction.
2019-07-16 17:48:07 +03:00
Joost Jager
8055bcf2e0
routing: report failures to mission control directly
As there is no more state kept in the payment session, failure reporting
can go straight to mission control.
2019-07-13 22:38:25 +02:00
Joost Jager
dc13da5abb
routing: move second chance logic into mission control
If nodes return a channel policy related failure, they may get a second
chance. Our graph may not be up to date. Previously this logic was
contained in the payment session.

This commit moves that into global mission control and thereby removes
the last mission control state that was kept on the payment level.

Because mission control is not aware of the relation between payment
attempts and payments, the second chance logic is no longer based
tracking second chances given per payment.

Instead a time based approach is used. If a node reports a policy
failure that prevents forwarding to its peer, it will get a second
chance. But it will get it only if the previous second chance was
long enough ago.

Also those second chances are no longer dependent on whether an
associated channel update is valid. It will get the second chance
regardless, to prevent creating a dependency between mission control and
the graph. This would interfer with (future) replay of history, because
the graph may not be the same anymore at that point.
2019-07-13 22:38:23 +02:00
Joost Jager
d31efddf1b
routing: move generateBandwidthHints 2019-07-13 22:38:21 +02:00
Joost Jager
37e2751695
routing+routerrpc: isolate payment session source from mission control 2019-07-13 22:38:19 +02:00
Joost Jager
cf3dd3fb94
routing/test: create mission control test context 2019-07-13 22:38:17 +02:00
nsa
7e6a34532e routing: add index map to distanceHeap
This commit adds the pubkeyIndices map to the distanceHeap to avoid
duplicate entries on the heap. This happened in the earlier iteration
of the findPath algorithm and would cause the driving loop to
evaluate already evaluated entries when there was no need.
2019-07-12 06:08:01 -04:00
nsa
61a1ab08fb routing+channeldb: findPath uses Vertex instead of LightningNode
This commit modifies the nodeWithDist struct to use a route.Vertex
instead of a *channeldb.LightningNode. This change, coupled with
the new ForEachNodeChannel function, allows the findPath Djikstra's
algorithm to cut down on database lookups since we no longer need
to call the FetchOtherNode function.
2019-07-12 06:08:01 -04:00
Joost Jager
2726f50d7c
htlcswitch: do not consider unknown failures an error 2019-07-11 19:49:49 +02:00
Joost Jager
418fe364de
htlcswitch: use distinct error for unreadable failures 2019-07-11 19:49:45 +02:00
Joost Jager
2b47632b26
htlcswitch+routing+routerrpc: return error source as index 2019-07-11 10:37:26 +02:00
Joost Jager
569f821142
routing/test: allow arbitrary graph source nodes 2019-07-11 10:37:22 +02:00
Olaoluwa Osuntokun
ba5fbb3c27
Merge pull request #3156 from joostjager/extended-fail
routerrpc: add more failure reasons and route hints
2019-07-08 19:12:03 -07:00
Joost Jager
5ce04091d8
routing+routerrpc+channeldb: return route on invalid payment details 2019-07-04 09:27:16 +02:00
Joost Jager
7bc56f5a7b
routing: add subscriber event constructors 2019-07-04 09:27:14 +02:00
Joost Jager
ae46fb00cb
routing+channeldb: add more failure reasons 2019-07-04 09:27:12 +02:00
nsa
678fb674eb routing: move call to PruneGraph outside of the collating loop
This commit moves the call to PruneGraph outside of the loop
that collates all of the spentOutputs. With this change, if
a node has been offline for a long period of time, resyncing
with the chain no longer takes up as much memory (1MB vs 200MB
in some cases) or time. Previously, PruneGraph was called
for every block and allocated a very large map further down
in the pruneGraphNodes function. Now, pruneGraphNodes is only
called once.
2019-07-02 21:21:28 -04:00
Johan T. Halseth
8b83f03308
rpcserver+router: log terminal sendPayment error 2019-07-01 16:22:57 -07:00
Conner Fromknecht
3a4aba463b
routing+watchtower/wtclient: prints stats once per minute 2019-06-19 20:07:11 -07:00
Olaoluwa Osuntokun
5f4accd39f
Merge pull request #3195 from halseth/routeing-nil-copy-pubkey
routing+zpay32: copy pubkeys before nilling Curve and spewing
2019-06-18 17:19:25 -07:00
Conner Fromknecht
529db69ec9
routing/router: log aggregate graph processing stats 2019-06-13 18:18:30 -07:00
Johan T. Halseth
386f8ece54
routing+zpay32: copy pubkeys before nilling Curve and spewing
Since nilling the pubkey curve will lead to a nil-pointer exception if
the key is later used for signature verification, we make sure to make a
copy before nilling and spewing.
2019-06-12 12:19:43 +02:00
Wilmer Paulino
c4546081c3
Merge pull request #3172 from joostjager/sendtoroute-timeout
routing+lnrpc: move default payment timeout out of router
2019-06-11 16:27:19 -07:00
Johan T. Halseth
5485101f9f
Merge pull request #3170 from cfromknecht/remove-router-reject-cache
routing/router: remove router-level reject cache
2019-06-11 21:52:27 +02:00
Joost Jager
2e920de292
routing+lnrpc: move default payment timeout out of router
This commit moves the default timeout out of router and thereby fixes a
bug that caused SendToRoute to not return the actual error, but a
timeout result instead. SendToRoute only tries a single route, so a
timeout should never happen.
2019-06-11 08:36:17 +02:00
Johan T. Halseth
2dea790b55
multi: make GetPaymentResult take payment hash
Used for logging in the switch, and when we remove the pending payments,
only the router will have the hash stored across restarts.
2019-06-07 16:53:32 +02:00
Conner Fromknecht
f8db49e7e4
routing/router: remove router-level reject cache
This is now redundant since we have one at the db-level
2019-06-06 13:02:16 +02:00
Joost Jager
07d289c14e
routing: add SendPaymentAsync to router 2019-06-05 12:41:51 +02:00
Joost Jager
eb2647e8fc
channeldb: add subscription to control tower
Allows other sub-systems to subscribe to payment success and fail
events.
2019-06-05 12:41:49 +02:00
Joost Jager
87d3207baf
channeldb+routing: move control tower interface to routing
This commit creates an empty shall for control tower in the routing
package. It is a preparation for adding event notification.
2019-06-05 12:41:47 +02:00
Joost Jager
19d5f8f82c
routing: move default cltv assignment out of router
This commit lifts default setting up to the rpc server level, in line
with other payment defaults.
2019-06-05 12:41:43 +02:00
Olaoluwa Osuntokun
7453dbeacc
Merge pull request #2802 from joostjager/probability
routing: probability based path finding
2019-06-05 11:30:55 +02:00
Joost Jager
054e42f680
routing+routerrpc: expose mission control parameters in lnd config
This commit exposes the three main parameters that influence mission
control and path finding to the user as command line or config file
flags. It allows for fine-tuning for optimal results.
2019-06-04 13:22:50 +02:00
Johan T. Halseth
6d52128da0
routing/router test: assert SendToRoute init values
This commit adds an assertion to the SendToRoute test that the payment
value stored to the DB during SendToRoute execution is the correct one.

This assertion would fail before the previous commit that fixed a
missing value initialization.
2019-06-04 10:34:59 +02:00
Johan T. Halseth
1161d87eec
routing/router: correct SendToRoute's amount record in DB
Previously we would mistakenly use the payment value from the dummy
LightningPayment struct, which would obviously be 0 always. Now we
instead calculate the value from the given route.
2019-06-04 10:13:33 +02:00
Joost Jager
f7982f0f4c
routing+routerrpc: expose mission control state over rpc
This commit exposes mission control state for rpc for development
purposes.
2019-06-04 10:00:27 +02:00
Joost Jager
7133f37bb8
routing: global probability based mission control
Previously every payment had its own local mission control state which
was in effect only for that payment. In this commit most of the local
state is removed and payments all tap into the global mission control
probability estimator.

Furthermore the decay time of pruned edges and nodes is extended, so
that observations about the network can better benefit future payment
processes.

Last, the probability function is transformed from a binary output to a
gradual curve, allowing for a better trade off between candidate routes.
2019-06-04 10:00:25 +02:00
Joost Jager
3349d517aa
routing: add min probability to path finding
This commit adds a new restriction to pathfinding that allows returning
only routes with a minimum success probability.
2019-06-04 10:00:23 +02:00
Joost Jager
6b70791c2d
routing: use probability source in path finding
This PR replaces the previously used edge and node ignore lists in path
finding by a probability based system. It modifies path finding so that
it not only compares routes on fee and time lock, but also takes route
success probability into account.

Allowing routes to be compared based on success probability is achieved
by introducing a 'virtual' cost of a payment attempt and using that to
translate probability into another cost factor.
2019-06-04 10:00:21 +02:00
Joost Jager
68f2a04f42
routerrpc: expose mission control reset rpc 2019-06-04 10:00:17 +02:00
Joost Jager
3db48dc53f
routing: Route String() method 2019-06-04 10:00:15 +02:00
Joost Jager
35003bd87c
routing/test: sort nodes for channel graph
This commit ensures that channel endpoints in EdgeInfo are ordered
node1 < node2 to not violate assumptions being made by dependent code.
2019-06-04 10:00:13 +02:00
Olaoluwa Osuntokun
4068e78af6
Merge pull request #2717 from cfromknecht/send-payment-payreq
lnrpc+rpcserver: display payreq via listpayments
2019-05-31 17:09:04 -07:00
Joost Jager
1b5655f7fa
Merge pull request #3113 from joostjager/sendtoroute-error-fix
routing: return structured error for send to route
2019-05-30 23:20:14 +02:00
Conner Fromknecht
078734b511
routing/router: set payreq before persisting PaymentCreationInfo 2019-05-29 16:31:12 -07:00
Joost Jager
cbfc7dd42d
routing: check findPath error 2019-05-29 14:07:00 +02:00
Joost Jager
0e273a5731
routing: return structured error for send to route 2019-05-29 14:06:58 +02:00
Johan T. Halseth
5882e54bf3
route test: add a simple unit test for the TotalFees method
Checks fees on 0, 1, and 2-hop routes.
2019-05-28 11:16:34 +02:00
Johan T. Halseth
f829d33d6f
route: return 0 fee from TotalFees for empty route.
Otherwise we would crash calling the method on empty routes.
2019-05-28 11:03:14 +02:00
Johan T. Halseth
649ee875fd
routing/router_test: add test case for ForwardingError on SendHTLC 2019-05-27 20:19:00 +02:00
Johan T. Halseth
1b788904f0
channeldb+router: record payment failure reason in db 2019-05-27 20:18:59 +02:00
Johan T. Halseth
60e2367973
routing/router_test: add TestRouterPaymentStateMachine
TestRouterPaymentStateMachine tests that the router interacts as
expected with the ControlTower during a payment lifecycle, such that it
payment attempts are not sent twice to the switch, and results are
handled after a restart.
2019-05-27 20:18:59 +02:00
Johan T. Halseth
3f76bc0629
routing+server: define PaymentSessionSource 2019-05-27 20:18:59 +02:00
Johan T. Halseth
f4306b1178
routing/payment_session+router: make PaymentSession interface 2019-05-27 20:18:58 +02:00
Johan T. Halseth
adc4640f4f
routing: export MissionControl 2019-05-27 20:18:58 +02:00
Johan T. Halseth
3323800e02
routing/router: resume payment state machine at startup
On startup the router will fetch the in-flight payments from the control
tower, and resume their execution.
2019-05-27 20:18:58 +02:00
Johan T. Halseth
677cb018c9
routing/control_tower: add FetchInFlightPayments 2019-05-27 20:18:58 +02:00
Johan T. Halseth
de1bf8a518
routing/router: persist payment state machine
This commit makes the router use the ControlTower to drive the payment
life cycle state machine, to keep track of active payments across
restarts.  This lets the router resume payments on startup, such that
their final results can be handled and stored when ready.
2019-05-27 20:18:58 +02:00
Johan T. Halseth
dd73c51a34
routing/payment_lifecycle: extract create attempt into method 2019-05-27 20:18:58 +02:00
Johan T. Halseth
e9b2182cdc
routing/payment_lifecycle: extract error handling into method 2019-05-27 20:18:58 +02:00
Johan T. Halseth
83bfaa4fb4
routing: extract payment flow into method on paymentLifecycle
This encapsulates all state needed to resume a payment from any point of
the payment flow, and that must be shared between the different stages
of the execution. This is done to prepare for breaking the send loop
into smaller parts, and being able to resume the payment from any point
from persistent state.
2019-05-27 20:18:57 +02:00
Johan T. Halseth
ae7bf2cb7b
routing/router: move sending and receiving payment result into loop 2019-05-27 20:18:57 +02:00
Wilmer Paulino
6e0084fec9
Merge pull request #3075 from cfromknecht/crtr-nil-path-curves
routing/router: nil path pubkeys before spewing
2019-05-23 20:01:54 -07:00
Xavi Soler
28021361d1
routing: add color to node update 2019-05-23 10:52:12 +02:00
Johan T. Halseth
2c1c29e060
Merge pull request #3081 from halseth/route-remove-totalfees
routing/route+multi: remove redundant TotalFees field
2019-05-20 15:00:01 +02:00
Johan T. Halseth
15bed506b1
routing/route+multi: remove redundant TotalFees field
Instead get it on demand using method TotalFees().
2019-05-16 23:58:32 +02:00
Johan T. Halseth
cd02c22977
htlcswitch+router: move deobfuscator creation to GetPaymentResult call
In this commit we move handing the deobfuscator from the router to the
switch from when the payment is initiated, to when the result is
queried.

We do this because only the router can recreate the deobfuscator after a
restart, and we are preparing for being able to handle results across
restarts.

Since the deobfuscator cannot be nil anymore, we can also get rid of
that special case.
2019-05-16 23:56:12 +02:00
Johan T. Halseth
5a8b892bb6
routing/router: extract sessionKey generation
We extract the session key such that we can later store it across
restarts.
2019-05-16 23:56:12 +02:00
Johan T. Halseth
ec087a9f73
htlcswitch+router: define PaymentResult, GetPaymentResult
This lets us distinguish an critical error from a actual payment result
(success or failure). This is important since we know that we can only
attempt another payment when a final result from the previous payment
attempt is received.
2019-05-16 23:56:12 +02:00
Johan T. Halseth
ed81cec0fa
routing/router: remove sendToSwitch method 2019-05-16 23:56:11 +02:00
Johan T. Halseth
c9e8ff6a34
switch+router+server: move NextPaymentID to router
This commit moves the responsibility of generating a unique payment ID
from the switch to the router. This will make it easier for the router
to keep track of which HTLCs were successfully forwarded onto the
network, as it can query the switch for existing HTLCs as long as the
paymentIDs are kept.

The router is expected to maintain a map from paymentHash->paymentID,
such that they can be replayed on restart. This also lets the router
check the status of a sent payment after a restart, by querying the
switch for the paymentID in question.
2019-05-16 23:56:06 +02:00
Johan T. Halseth
f1cb54f943
routing/router: define PaymentAttemptDispatcher interface
The switch satisfies this interface, and makes it easy to mock the send
method from the router.
2019-05-16 23:53:08 +02:00
Joost Jager
ba3fa94268
lnrpc+routing: Only accept a single route for SendToRoute 2019-05-15 11:54:46 +02:00
Conner Fromknecht
4d26e90ef3
routing/router: nil path pubkeys before spewing 2019-05-14 20:32:58 -07:00
Olaoluwa Osuntokun
f8c824fb1d
Merge pull request #3054 from joostjager/remove-k-shortest
lnrpc+routing: remove k shortest path finding
2019-05-10 15:51:01 -07:00
Johan T. Halseth
10070ecab7
lnwallet: make BlockChainIO.GetUTXO take cancel chan
Use quit channels as cancel chan for call to GetUTXO.
2019-05-09 14:44:47 +02:00
Joost Jager
7a5bd29a69
lnrpc+routing: remove k shortest path finding 2019-05-07 19:54:06 +02:00
Johan T. Halseth
c4415f0400
Merge pull request #3044 from cfromknecht/spelling-fixes
multi: fix spelling mistakes
2019-05-07 08:50:36 +02:00
Conner Fromknecht
3f59e31248
routing/router: break rescans on router quit
This commit reevaluates the router's quit channel between each block
during the initial call to syncGraphWithChain, which, in the worst case,
may have to scan several thousand blocks on startup if the node has not
been active for some time. Without this, attempting to stop the daemon
will not exit until the rescan has completed, which for certain backends
could be several hours.
2019-05-06 15:04:40 -07:00
Conner Fromknecht
17b2140cb5
multi: fix spelling mistakes 2019-05-04 15:35:37 -07:00
Olaoluwa Osuntokun
bab957382f
router: convert Route.ToHopPayloads() to Route.ToSphinxPath()
In this commit, we update the process that we use to generate a sphinx
packet to send our onion routed HTLC. Due to recent changes in the
`sphinx` package we use, we now need to use a new PaymentPath struct. As
a result, it no longer makes sense to split up the nodes in a route and
their per hop paylods as they're now in the same struct. All tests have
been updated accordingly.
2019-04-30 20:13:32 -07:00
Johan T. Halseth
ee257fd0eb
multi: move Route to sub-pkg routing/route 2019-04-29 14:52:33 +02:00
Wilmer Paulino
998680ad59
routing: fall back to edge MaxHTLC within findPath instead of Capacity
In this commit, we make our findPath function use an edge's MaxHTLC as
its available bandwidth instead of its Capacity. We do this as it's
possible for the capacity of an edge to not exist when operating as a
light client. For channels that do not support the MaxHTLC optional
field, we'll fall back to using the edge's Capacity.
2019-04-18 21:57:42 -07:00
Wilmer Paulino
0b0a9f4172
channeldb+routing: refactor DeleteChannelEdge to use ChannelID
In this commit, we refactor DeleteChannelEdge to use ChannelIDs rather
than ChannelPoints. We do this as the only use of DeleteChannelEdge is
when we are pruning zombie channels from our graph. When running under a
light client, we are unable to obtain the ChannelPoint of each edge due
to the expensive operations required to do so. As a stop-gap, we'll
resort towards using an edge's ChannelID instead, which is already
gossiped between nodes.
2019-04-18 21:57:41 -07:00
Wilmer Paulino
5d3621cc83
routing: skip max htlc validation when capacity is 0
Since light clients no longer have access to an edge's capacity, they
are unable to validate whether the max HTLC value for an updated edge
policy respects the capacity limit. As a stop-gap, we'll skip this
check.
2019-04-18 21:57:38 -07:00
Wilmer Paulino
f2637d63ba
routing: prevent fetching blocks for graph edges with AssumeChannelValid
This serves as a stop-gap for light clients as blocks need to be
downloaded from the P2P network, and even with caches, would be too
costly for them to verify. Doing this has two side effects however:
we'll no longer know of the channel capacity and outpoint, which are
essential for some of lnd's responsibilities.
2019-04-18 21:57:37 -07:00
Wilmer Paulino
fd1aa478a9
routing: disable FilteredChainView when AssumeChannelValid is active
In this commit, we disable attempting to determine when a channel has
been closed out on-chain whenever AssumeChannelValid is active. Since
the flag indicates that performing this operation is expensive, we do
this as a temporary optimization until we can include proofs of channels
being closed in the gossip protocol.

With this change, the only way for channels being removed from the graph
will be once they're considered zombies: which can happen when both
edges of a channel have their disabled bits set or when both edges
haven't had an update within the past two weeks.
2019-04-18 21:57:37 -07:00
Wilmer Paulino
7e7b8a1940
routing: prune graph nodes after pruning zombie channels
We do this to ensure we don't leave any stray nodes in our graph that
were part of the zombie channels that we've pruned.
2019-04-18 21:57:36 -07:00
Wilmer Paulino
7eb720e535
routing: mark policy as stale if edge remains double disabled
To ensure we don't mark an edge as live again just because an update
with a fresh timestamp was received, we'll ensure that we reject any
new updates for zombie channels if they remain disabled when running
with AssumeChannelValid.
2019-04-18 21:57:35 -07:00
Wilmer Paulino
f23c3b488e
routing: prune channels with disabled bit set on both edges
In this commit, we add an additional heuristic when running with
AssumeChannelValid. Since AssumeChannelValid being present assumes that
we're not able to quickly determine whether channels are valid, we also
assume that any channels with the disabled bit set on both sides are
considered zombie. This should be relatively safe to do, since the
disabled bits are usually set when the channel is closed on-chain. In
the case that they aren't, we'll have to wait until both edges haven't
had a new update within two weeks to prune them.
2019-04-18 21:57:34 -07:00
Wilmer Paulino
78bdcbb115
routing: prune channels only if both edges are present
We do this to ensure we don't prune too aggressively, as it's possible
that we've only received the channel announcement for a channel, but not
its accompanying channel updates.
2019-04-18 21:57:31 -07:00
Wilmer Paulino
29664c9704
routing: make test channel policies optional 2019-04-18 20:47:32 -07:00
Olaoluwa Osuntokun
373927fb3e
Merge pull request #2950 from joostjager/remove-route-cache
routing: remove route cache
2019-04-17 19:34:19 -07:00
Olaoluwa Osuntokun
a3c980b1e8
Merge pull request #2871 from halseth/router-subscribetopology-check-started
routing/notifications: check router started on SubscribeTopology
2019-04-16 21:32:10 -07:00
Johan T. Halseth
2b9efcbaeb
routing/router: define ErrRouterShuttingDown 2019-04-15 16:55:10 +02:00