This way we can greatly simplify the method signatures, also paving the
upcoming changes where we wanna make it clear when updating the
monitorRecord, we only touch a portion of it.
This commit shortens the function signature of `storeRecord`, also makes
sure we don't call `t.records.Store` directly but always using
`storeRecord` instead so it's easier to trace the record creation.
When mining lots of blocks in the itest, the subsystems can be out of
sync in terms of the best block height. We now assert the num of pending
sweeps on Alice's node to give her more time to sync the blocks,
essentially behaving like a `time.Sleep` as in reality, the blocks would
never be generated this fast.
We need to make sure Caarol finishes settling her invoice with Bob
before shutting down, so we make sure `AssertHTLCNotActive` on Bob
happens before shutting node Carol.
We add a new config option to set the `ProofMatureDelta` so the users
can tune their graphs based on their own preference over the num of
confs found in the announcement signatures.
The mocked peer used here blocks on `sendToPeer`, which is not the
behavior of the `SendMessageLazy` of `lnpeer.Peer`. To reflect the
reality, we now make sure the `sendToPeer` is non-blocking in the tests.
This commit fixes the following race,
1. syncer(state=syncingChans) sends QueryChannelRange
2. remote peer replies ReplyChannelRange
3. ProcessQueryMsg fails to process the remote peer's msg as its state
is neither waitingQueryChanReply nor waitingQueryRangeReply.
4. syncer marks its new state waitingQueryChanReply, but too late.
The historical sync will now fail, and the syncer will be stuck at this
state. What's worse is it cannot forward channel announcements to other
connected peers now as it will skip the broadcasting during initial
graph sync.
This is now fixed to make sure the following two steps are atomic,
1. syncer(state=syncingChans) sends QueryChannelRange
2. syncer marks its new state waitingQueryChanReply.
We sometimes see `timeout waiting for UTXOs` error from bitcoind-related
itests due to the chain backend not synced to the miner. We now assert
it's synced before continue.
This commit removes the panic used in checking the shutdown log.
Instead, the error is returned and asserted in `shutdownAllNodes` so
it's easier to check which node failed in which test. We also catch all
the errors returned from `StopDaemon` call to properly access the
shutdown behavior.
Previous splitting logic simply put all the remainder in the last
tranche, which could make the last tranche run significantly more test
cases. We now change it so the remainder is evened out across tranches.
Most of the time we only need to fund the node with given number of
UTXOs without concerning the amount, so we add the more efficient
funding method as it mines a single block in the end.
To make the CI indicative, we now starting tracking the flaky tests
found when running on Windows. As a starting point, rather than ignore
the windows CI entirely, we now identify there are cases where lnd can
be buggy when running in windows.
We should fix the tests in the future, otherwise the windows build
should be deleted.
This has been seen in the itest which can lead to the node startup
failure,
```
2024-11-20 18:55:15.727 [INF] RPCS: Max websocket clients exceeded [25] - disconnecting client 127.0.0.1:57224
```
We need to mine an empty block as the tx may already have entered the
mempool. This should be fixed once we start using the sweeper to handle
the justice tx.
Soemtimes the node may be hanging for a while without being noticed,
causing failures in its following tests, thus making the debugging
extrememly difficult. We now assert the node has been shut down from the
logs to assert the shutdown process behaves as expected.
Since we don't have standby nodes anymore, we don't need to close the
channels when the test finishes. Previously we would do so to make sure
the standby nodes have a clean state for the next test case, which is no
longer relevant.
This commit removes the usage of the standby nodes and uses
`CreateSimpleNetwork` when applicable. Also introduces a helper method
`NewNodeWithCoins` to quickly start a node with funds.
Prepares the upcoming refactor. We now never call `ht.Alice` directly,
instead, we always init `alice := ht.Alice` so it's easier to see how
they are removed in a following commit.
This commit adds a new flag to shuffle all the test cases before running
them so tests which require lots of blocks to be mined are less likely
to be run in the same tranch.
The other benefit is this approach provides a more efficient way to
figure which tests are broken since all the differnet backends are
running different tranches in their builds, we can identify more failed
tests in one push.
The method `AssertActiveHtlcs` is now removed due to it's easy to be
misused. To assert a given htlc, use `AssertOutgoingHTLCActive` and
`AssertIncomingHTLCActive` instead for ensuring the HTLC exists in the
right direction. Although often the case `AssertNumActiveHtlcs` would be
enough as it implicitly checks the forwarding behavior for an
intermediate node by asserting there are always num_payment*2 HTLCs.
Yong