If a channel goes offline while the count of outstanding outgoing HTLCs
exceeds the limit that we enforce against the peer, then the channel
could never be brought online again because `add_htlc` called by
`channel_force_htlcs` in `channeld/full_channel.c` would return
`CHANNEL_ERR_TOO_MANY_HTLCS`. The protocol specification actually does
allow us to exceed the limits that we are enforcing against the peer;
we are only prohibited from exceeding the limits that the peer is
enforcing against us. `add_htlc` takes an `enforce_aggregate_limits`
parameter that appears to have been intended for `channel_force_htlcs`
to exempt the local node from obeying the limits that it is enforcing
against the peer, but this parameter was only being respected for the
total HTLC value-in-flight check but not for the HTLC count check. This
commit respects the parameter for the HTLC count check as well and
resolves the problem of "Could not restore HTLCs".
Fixes: #5636
Changelog-Fixed: channeld: Channel reinitialization no longer fails when the number of outstanding outgoing HTLCs exceeds `max_accepted_htlcs`.
"sphinx_add_hop" takes a literal hop to include,
"sphinx_add_modern_hop" prepends the length. Now we always prepend a
length, make it clear that the literal version is a shortcut:
* sphinx_add_hop -> sphinx_add_hop_has_length
* sphinx_add_modern_hop -> sphinx_add_hop
In addition, we check that length is actually correct! This means
`createonion` can no longer create legacy or otherwise-invalid onions:
fix tests and update man page to remove legacy usage.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Changed: JSON-RPC: `createonion` no longer allows non-TLV-style payloads.
We still have an "enum forward_style" for the database, where old-style
forwards can still exist.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Removed: Protocol: we no longer forward HTLCs with legacy onions.
We used to ensure the l3<->l4 channel was private by
simply not mining enough blocks to announce. Then
we started mining 13 blocks to close the channel, and
it will get announced, causing a failure:
```
assert non_public(l2) == []
> wait_for(lambda: non_public(l3) == [scid34, scid34])
...
> raise ValueError("Timeout while waiting for {}", success)
E ValueError: ('Timeout while waiting for {}', <function test_gossip_persistence.<locals>.<lambda> at 0x7f6cc69b4170>)
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Second pay can fail if first is not completely settled:
```
def test_zeroconf_forward(node_factory, bitcoind):
"""Ensure that we can use zeroconf channels in forwards.
...
# Make sure (esp in non-dev-mode) blockheights agree so we don't WIRE_EXPIRY_TOO_SOON...
sync_blockheight(bitcoind, [l1, l2, l3])
inv = l3.rpc.invoice(42 * 10**6, 'inv1', 'desc')['bolt11']
l1.rpc.pay(inv)
# And now try the other way around: zeroconf channel first
# followed by a public one.
wait_for(lambda: len(l3.rpc.listchannels()['channels']) == 4)
inv = l1.rpc.invoice(42, 'back1', 'desc')['bolt11']
> l3.rpc.pay(inv)
...
> raise RpcError(method, payload, resp['error'])
E pyln.client.lightning.RpcError: RPC call failed: method: pay, payload: {'bolt11': 'lnbcrt420p1p3junrssp588gtjzmlrr4pfj7ssmdlulzhhushrpq3rdqxjuz2m33scsvzdjlspp5fk5yhu6netc0d0sgp8es52vjk6akhd3uayr08u8max4d8rwzpjuqdq8v3jhxccxqyjw5qcqp99qyysgqcpyyugejv4nya97v6gw8fhtr0ru3vq87jjlltav99wlat436a95n0z8yzdp699p9md0zz9tmnsjpvfj622n9g9fh7r6ldhpgh9wmr4qpcru3rk'}, error: {'code': 210, 'message': 'Destination 0266e4598d1d3c415f572a8488830b60f7e744ed9235eb0b1ba93283b315c03518 is not reachable directly and all routehints were unusable.', 'attempts': [{'status': 'failed', 'failreason': 'Destination 0266e4598d1d3c415f572a8488830b60f7e744ed9235eb0b1ba93283b315c03518 is not reachable directly and all routehints were unusable.', 'partid': 0, 'amount_msat': 42msat}]}
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We have to allow them (as otherwise `fees_collected_msat` in getinfo breaks),
but it means that actually, in_htlc_id might be missing in listforwards
(also, out_htlc_id might be missing, which we didn't catch before).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Fixes: #5628
This is just before the introduction of `get_json_id`, but has the
correct dependency constraints such that all packages can be updated
to >=v0.12 and we don't mix minor versions.
Otherwise what the hook sees is actually a lie, and if it sets it
we might override it.
The side effect is that we add an explicit "forward_to" field, and
allow hooks to override it. This lets a *hook* control channel
choice explicitly.
Changelod-Added: Plugins: `htlc_accepted_hook` return can specify what channel to forward htlc to.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is the minimal change to meet the desired outcome of https://github.com/lightning/bolts/issues/934
which wants to give obsolete-db nodes a chance to fix things up, before we
close the channel.
We need to dance around a bit here, since we *will* close the channel if
we receive an ERROR, so we suppress that.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We needed to clone a lot since we were using a non-mutable borrow of
the cln-rpc version to then create the cln-grpc version. This alone
reduces the number of allocations of the `listpeers` test from 52
allocations to 38.
We filter based on the environment variable `CLN_PLUGIN_LOG`,
defaulting to `info` as that is not as noisy as `debug` or `trace`, at
least libraries will not spam us too heavily.
Changelog-Added cln-plugin: The logs level from cln-plugins can be configured by the `CLN_PLUGIN_LOG` environment variable.
We had a bit of a chicken-and-egg problem, where we instantiated the
`state` to be managed by the `Plugin` during the very first step when
creating the `Builder`, but then the state might depend on the
configuration we only get later. This would force developers to add
placeholders in the form of `Option` into the state, when really
they'd never be none after configuring.
This defers the binding until after we get the configuration and
cleans up the semantics:
- `Builder`: declare options, hooks, etc
- `ConfiguredPlugin`: we have exchanged the handshake with
`lightningd`, now we can construct the `state` accordingly
- `Plugin`: Running instance of the plugin
Changelog-Changed: cln-plugin: Moved the state binding to the plugin until after the configuration step
This adds a new "chan_dying" message to the gossip_store, but since we
already changed the minor version in this PR, we don't bump it again.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: Protocol: We now delay forgetting funding-spent channels for 12 blocks (as per latest BOLTs, to support splicing in future).
It's a bit more optimal, and tells gossipd exactly what height the
spend occurred at (with multiple blocks, it's not always the current
height!). It will need that next patch.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: Protocol: We now set the `dont_forward` bit on private channel_update's message_flags (as per latest BOLTs).
We will now simply reject old-style ones as invalid. Turns out the
only trace we could find is a channel between two nodes unconnected to
the rest of the network.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Changed: Protocol: We now require all channel_update messages include htlc_maximum_msat (as per latest BOLTs)
Many changes to gossmap (including the pending ones!) don't actually
concern readers, as long as they obey certain rules:
1. Ignore unknown messages.
2. Treat all 16 upper bits of length as flags, ignore unknown ones.
So now we split the version byte into MAJOR and MINOR, and you can
ignore MINOR changes.
We don't expose the internal version (for creating the map)
programmatically: you should really hardcode what major version you
understand!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If they really upgrade directly from 0.9.2, it will simply delete the
store and re-fetch it.
We still update from v9 (which could be v0.11), since it's a noop.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If channeld hasn't exited yet, it's possible we'll send the message (we would fail later, in
waitsendpay, but just not immediately). So wait for that explicitly.
```
2022-09-22T22:49:59.6737296Z with pytest.raises(RpcError, match=r'failed: WIRE_TEMPORARY_CHANNEL_FAILURE \(First peer not ready\)'):
2022-09-22T22:49:59.6737566Z > l1.rpc.sendpay(route, rhash, payment_secret=inv['payment_secret'])
2022-09-22T22:49:59.6737865Z E Failed: DID NOT RAISE <class 'pyln.client.lightning.RpcError'>
2022-09-22T22:49:59.6737873Z
```
And from the listpeers output, ou can see "connected" false, but owner channeld:
```
2022-09-22T22:49:59.7493163Z DEBUG:root:{
2022-09-22T22:49:59.7493320Z "id": "-c:listpeers#26",
2022-09-22T22:49:59.7493397Z "result": {
2022-09-22T22:49:59.7493477Z "peers": [
2022-09-22T22:49:59.7493548Z {
2022-09-22T22:49:59.7493709Z "id": "022d223620a359a47ff7f7ac447c85c46c923da53389221a0054c11c1e3ca31d59",
2022-09-22T22:49:59.7493801Z "connected": false,
2022-09-22T22:49:59.7493884Z "channels": [
2022-09-22T22:49:59.7493955Z {
2022-09-22T22:49:59.7494058Z "state": "CHANNELD_NORMAL",
2022-09-22T22:49:59.7494250Z "scratch_txid": "4b95a3b1b5e1a970401a169a3697f3a9bfbfbcb59d3d21434aa1f3fb2980db8d",
2022-09-22T22:49:59.7494365Z "last_tx_fee_msat": "7965000msat",
2022-09-22T22:49:59.7494437Z "feerate": {
2022-09-22T22:49:59.7494529Z "perkw": 11000,
2022-09-22T22:49:59.7494618Z "perkb": 44000
2022-09-22T22:49:59.7494690Z },
2022-09-22T22:49:59.7494785Z "owner": "channeld",
2022-09-22T22:49:59.7494894Z "short_channel_id": "103x1x0",
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
In particular, some field names have changed. lnprototest in master
has already upgraded names, so this brings us into compatibility again.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Changed: pyln-spec: package updated to latest spec version.
1. Update default specdir to the new modern name.
2. Don't use python to extract version, use sed.
3. Use poetry publish.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
CI is really slow: it sees all three expire at once. But making the
timeouts too long is painful in non-VALGRIND, so I ended up making it
conditional.
```
# First it expires.
wait_for(lambda: only_one(l3.rpc.listinvoices('inv1')['invoices'])['status'] == 'expired')
# Now will get autocleaned
wait_for(lambda: l3.rpc.listinvoices('inv1')['invoices'] == [])
> assert l3.rpc.autoclean_status()['autoclean']['expiredinvoices']['cleaned'] == 1
E assert 3 == 1
tests/test_plugin.py:2975: AssertionError
```
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Christian Decker
Justin Litchfield
Matt Whitlock
Rusty Russell
joemphilips
elsirion
Melroy van den Berg
Vincenzo Palazzo
Greg Sanders