- the fee negotiation is split into smaller functions, reducing the scope of variables.
- the while loop logic is condensed in a few lines, so it is easier to understand termination conditions.
- removed code that was never executed
Messages are sent in reestablish_channel (async)
Message checks and force_close are performed in on_channel_reestablish (not async).
That task should not be cancelled if the connection is closed.
Revert 57583c05cf
* Refactor `create_trampoline_route`.
* Enables end-to-end multi-trampoline multipart payments.
Trampoline-to-legacy payments are still not enabled, as this is
currently not supported by Eclair.
* Reverts to a global trampoline fee level, as trampoline failures
are currently not handled properly, see (#7648), which doubles
fee rates.
In particular, in the regtests, with incoming peers, we can have multiple transports open with the same node simultaneously
(see e.g. lnworker._request_force_close_from_backup).
We now use the first few bytes of peer_pubkey, as that is potentially familiar to users,
and the first few bytes of sha256(id(self)) to mitigate collisions in case the peer_pubkeys collide.
log excerpt:
```
I/P | lnpeer.Peer.[LNWallet, 030f0bf260-e0b33756] | handshake done for 030f0bf260acdbd3edcad84d7588ec7c5df4711e87e6a23016f989b8d3a4147230@163.172.94.64:9735
D/P | lnpeer.Peer.[LNWallet, 030f0bf260-e0b33756] | Sending INIT
I/P | lnpeer.Peer.[LNWallet, 03933884aa-5e5dce45] | handshake done for 03933884aaf1d6b108397e5efe5c86bcf2d8ca8d2f700eda99db9214fc2712b134@34.250.234.192:9735
D/P | lnpeer.Peer.[LNWallet, 03933884aa-5e5dce45] | Sending INIT
D/P | lnpeer.Peer.[LNWallet, 030f0bf260-e0b33756] | Received INIT
I/P | lnpeer.Peer.[LNWallet, 02651acf4a-79696c42] | handshake done for 02651acf4a7096091bf42baad19b3643ea318d6979f6dcc16ebaec43d5b0f4baf2@82.119.233.36:19735
D/P | lnpeer.Peer.[LNWallet, 02651acf4a-79696c42] | Sending INIT
D/P | lnpeer.Peer.[LNWallet, 03933884aa-5e5dce45] | Received INIT
I/P | lnpeer.Peer.[LNWallet, 030f0bf260-e0b33756] | saved remote_update
D/P | lnpeer.Peer.[LNWallet, 030f0bf260-e0b33756] | Received CHANNEL_REESTABLISH
```
aiorpcx 0.20 changed the behaviour/API of TaskGroups.
When used as a context manager, TaskGroups no longer propagate
exceptions raised by their tasks. Instead, the calling code has
to explicitly check the results of tasks and decide whether to re-raise
any exceptions.
This is a significant change, and so this commit introduces "OldTaskGroup",
which should behave as the TaskGroup class of old aiorpcx. All existing
usages of TaskGroup are replaced with OldTaskGroup.
closes https://github.com/spesmilo/electrum/issues/7446
follow-up https://github.com/spesmilo/electrum/pull/7202
defaultdict[int] is a type!
```
>>> from collections import defaultdict
>>> d = defaultdict[int]
>>> d[2]
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
TypeError: There are no type variables left in collections.defaultdict[int]
```
Also, prior to py3.9, it is a TypeError.
Alice sends and HTLC: Alice->Carol->Dave
we need a lot of messages back and forth to happen:
- Alice adds HTLC to chan_AC, sends sig, Carol revacks, sends sig, Alice revacks;
- only then Carol adds HTLC to chan_CD, sends sig, Dave revacks, sends sig, Carol revacks
on CI, 0.5 seconds is often not enough for this it seems.
- Separates the trampoline and local routing multi-part payment cases.
- Ask only for splits that don't send over a single channel (those have
been tried already in the single-part case).
- Makes sure that create_routes_for_payment only yields partial routes
that belong to a single split configuration.
- Tracks trampoline fee levels on a per node basis, previously, in the
case of having two channels with a trampoline forwarder, the global
fee level would have increased by two levels upon first try.
- better error handling: previously we stopped all attempts on any of
TRAMPOLINE_EXPIRY_TOO_SOON, UNKNOWN_NEXT_PEER, TEMPORARY_NODE_FAILURE.
Instead we should retry (but see code comments).
- previously payments failed if ALL of the following criteria applied:
- sender is paying via trampoline, but not via the ACINQ node (which is
special cased)
- receiver only has private channels and has put r_tags into invoice, along
with setting the trampoline feature bit in the invoice, however the receiver
is not connected to any trampoline forwarders directly
The sender would then assume that the private routing hints in the invoice
correspond to trampoline forwarders.
- also, previously if both the sender and the recipient used trampoline and
they shared a trampoline forwarder (that they were both connected to), the
private channels the recipient had (with nodes other than the shared TF)
would never be attempted.
Adds liquidity hints for the sending capabilities of routing channels in the
graph. The channel blacklist is incorporated into liquidity hints.
Liquidity hints are updated when a payment fails with a temporary
channel failure or when it succeeds. Liquidity hints are used to give a
penalty in the _edge_cost heuristics used by the pathfinding algorithm.
The base penalty in (_edge_cost) is removed because it is now part of the
liquidity penalty. We don't return early from get_distances, as we want
to explore all channels.
The test framework did not like the "Test" prefix.
```
electrum/tests/test_lnpeer.py:305
/home/travis/build/spesmilo/electrum/electrum/tests/test_lnpeer.py:305: PytestCollectionWarning: cannot collect test class 'TestSuccess' because it has a __init__ constructor (from: electrum/tests/test_lnpeer.py)
```
When forwarding an HTLC and the outgoing channel doesn't have
enough funds or other transient issues, we are failing the
HTLC with a temporary channel failure.
From BOLT #04:
* if during forwarding to its receiving peer, an otherwise unspecified,
transient error occurs in the outgoing channel (e.g. channel capacity
reached, too many in-flight HTLCs, etc.):
-> return a temporary_channel_failure error.
This is most useful when receiving MPP where there is a non-trivial chance
that we have received some HTLCs for a payment but not all, and the user
closes the program. We try to fail them and wait for the fails to get
ACKed, with a timeout of course.
similar to 05fd424548
from logs when running tests:
--- Logging error ---
Traceback (most recent call last):
File "...\Python39\lib\logging\__init__.py", line 1082, in emit
stream.write(msg + self.terminator)
ValueError: I/O operation on closed file.
Call stack:
File "...\Python39\lib\threading.py", line 912, in _bootstrap
self._bootstrap_inner()
File "...\Python39\lib\threading.py", line 954, in _bootstrap_inner
self.run()
File "...\Python39\lib\threading.py", line 892, in run
self._target(*self._args, **self._kwargs)
File "...\electrum\electrum\sql_db.py", line 71, in run_sql
self.logger.info("SQL thread terminated")
Message: 'SQL thread terminated'
Arguments: ()
The test failures corresponding to single-part (non-MPP) payments expose a bug.
see 196b4c00a3/electrum/lnpeer.py (L1538-L1539)
`lnworker.add_received_htlc` is not called for single-part payments...
