Integration Tests Fail Locally: Bitcoind Wallet Init Errors & RPC Timeouts #1180

Okay... So on my local setup I’ve been consistently seeing integration tests fail when running ./contrib/test.sh. Unit tests complete successfully (161 passed), but as soon as the integration suite starts, every case that spins up our embedded bitcoind instance blows up with errors like:

• Error: it appears that bitcoind is not reachable: Could not create or load wallet
• JsonRpc(Transport(Minreq(IoError(Custom { kind: TimedOut, error: "the timeout of the request was reached" }))))
• repeated suggestions from bitcoind to restart with -reindex or -reindex-chainstate

This suggests bitcoind either can’t start or dies before the RPC client connects, which then causes the v1/v2 and batching tests to panic (timeout/fatal).

At the moment, I’m assuming this is something peculiar to my environment—maybe someone else has the same issues when they do the following:

• Run: dart test test/test_payjoin_integration_test.dart -p vm --plain-name 'Test integration Test integration v2 to v2' --reporter expanded
  Result: Fails after ~30 s with TimeoutException … Test timed out after 30 seconds while the OHTTP relay is listening (e.g. OHTTP relay binding to port [::]:52718). No additional stack trace beyond the harness suggesting --chain-stack-traces.

• Run: ./contrib/test.sh (run from repo root)
  Result: Unit tests pass, but the integration binary trips over every scenario that needs bitcoind. Each failing case reports JsonRpc(Transport(Minreq(IoError(Custom { kind: TimedOut, error: "the timeout of the request was reached" })))), and the panics point to payjoin/tests/integration.rs (e.g. line 492 for v2_to_v2). Only integration::v2::test_bad_ohttp_keys succeeds; the remaining 12 integration tests fail with the same RPC timeout.

I'm on MacOS 15.7.1, Intel i9