Messages sent as part of the payment were getting interleaved with the channel_reestablish.
It does not actually make sense to do a payment and then reestablish the channel in the same transport -- the channel is supposed to already have been reestablished to do a payment in the first place.
So, after payment, strip down the transport, and set up a new transport before reestablishing.
Traceback (most recent call last):
File "...\Python\Python38\lib\unittest\case.py", line 60, in testPartExecutor
yield
File "...\Python\Python38\lib\unittest\case.py", line 676, in run
self._callTestMethod(testMethod)
File "...\Python\Python38\lib\unittest\case.py", line 633, in _callTestMethod
method()
File "...\electrum\electrum\tests\test_lnpeer.py", line 262, in test_reestablish_with_old_state
run(f())
File "...\electrum\electrum\tests\test_lnpeer.py", line 302, in run
return asyncio.run_coroutine_threadsafe(coro, loop=asyncio.get_event_loop()).result()
File "...\Python\Python38\lib\concurrent\futures\_base.py", line 439, in result
return self.__get_result()
File "...\Python\Python38\lib\concurrent\futures\_base.py", line 388, in __get_result
raise self._exception
File "...\electrum\electrum\tests\test_lnpeer.py", line 260, in f
await gath
File "...\electrum\electrum\lnpeer.py", line 439, in _message_loop
self.process_message(msg)
File "...\electrum\electrum\lnpeer.py", line 159, in process_message
execution_result = f(payload)
File "...\electrum\electrum\lnpeer.py", line 1308, in on_revoke_and_ack
chan.receive_revocation(rev)
File "...\electrum\electrum\lnchannel.py", line 556, in receive_revocation
raise Exception('revoked secret not for current point')
Exception: revoked secret not for current point
Previously we would put fake chan announcement and fake outgoing chan upd
for own channels into db (to make path finding work). See Peer.add_own_channel().
Now, instead of above, we pass a "my_channels" param to the relevant ChannelDB methods.
- persisted states are saved
- state transitions are checked
- transient states are stored in channel.peer_state
- new channel states: 'PREOPENING', 'FUNDED' and 'REDEEMED'
- upgrade storage to version 21
- Move 'handle_error_code_from_failed_htlc' to channel_db,
and call it from pay_to_route, because it should not be
called when HTLCs are forwarded.
- Replace 'payment_received' and 'payment_status'
callbacks with 'invoice_status' and 'request_status'.
- Show payment error logs in the Qt GUI
- In the invoices list, show paid invoices for which
we still have the log.
The few other cases that used SimpleConfig.get_instance() now
either get passed a config instance, or they try to get a reference
to something else that has a reference to a config.
(see lnsweep, qt/qrcodewidget, qt/qrtextedit)
- use InvoiceInfo (NamedTuple) for normal operations,
because lndecode operations can be very slow.
- all invoices/requests are stored in wallet
- invoice expiration detection is performed in wallet
- CLI commands: list_invoices, add_request, add_lightning_request
- revert 0062c6d695 because it forbids self-payments
- commands are async
- the asyncio loop is started and stopped from the main script
- the daemon's main loop runs in the main thread
- use jsonrpcserver and jsonrpcclient instead of jsonrpclib
- use gossip_queries_req instead of initial_routing_sync
- add connected nodes to recent peers only after successful init
- derive timestamp used with gossip_timestamp_filter from channel_db
- fix query_short_channel_ids:
1. channel IDs must be sorted with zlib
2. limit request to 100
3. do not abuse this to request node_announcements; it is fine not to have all nodes.
- fix get_recent_peers:
1. do not set last_connected_date to 'now' if we never connected a node
2. sql query was misconstructed and was returning only one peer
- populate FALLBACK_NODE_LIST_MAINNET with nodes that have the requested flags
