Block-based deposit fetching from the internal lnd wallet was
susceptible to wallet syncing issues. Replace it with interval-based
polling. Reconciliation errors are now logged instead of being fatal,
improving resilience during transient failures.
Remove unreachable error check after filterNewDeposits which does not
return an error. The err variable was already handled from the
ListUnspent call above and could never be non-nil at this point.
Add macaroon permissions for the new StaticOpenChannel RPC method
and change StaticAddressLoopIn action from read to execute to match
other swap operations.
In this commit we import the lnd's lightning.proto into client.proto to
then be able to use lnrpc.OutPoint and lnrpc.OpenChannelRequest instead
of the looprpc.OutPoint type.
This is safe because the lnrpc and looprpc versions of OutPoint are the
same type.
In successive commits we want to access lnrpc specifc methods
and objects, so here we import github.com/lightningnetwork/lnd v0.20.1-beta
into looprpc and update relevant versions.
- Change return signature of startNewLoopIn from (*swapConfig, error, *loopInSwap) to (*swapConfig, *loopInSwap, error)
- Fixes ST1008: error should be returned as the last argument
- Update assets/client.go to use grpc.NewClient
- Update swap_server_client.go to use grpc.NewClient
- Replace grpc.WithInsecure() with insecure.NewCredentials()
- reset htlcConfirmed when the HTLC conf subscription errors and we re-register
- add regression test ensuring re-registers after a conf error require a fresh
confirmation instead of sweeping on a stale one
In order for the LiT CI jobs to use the `swapserverrpc` protos that
exist on the latest `master` version of the `loop` repo, we need to
update the `lit-setup` action to replace the litd `swapserverrpc`
dependency with the latest version that exists in the `loop` repo.
Fixes a crash window where handleConf updated the batch row to confirmed but
failed before marking sweeps complete, so re-added sweeps spawned a duplicate
batch or kept retrying. Batch confirmation and sweep completion are now
persisted inside a single DB transaction ConfirmBatchWithSweeps, and handleConf
uses the helper to atomically store the batch and the set of confirmed sweeps.
Added TestSweepBatcherConfirmedBatchIncompleteSweeps that runs against the real
loopdb backend, injects a failure mid-transaction, and verifies the database
never ends up with confirmed=true batches paired with completed=false sweeps.
