Technical Debt: Wallet Layer & Network 'listen' testing

[KtorZ]

Issue Number

#94

Overview

• I have removed unused code from 'watchWallet' in the wallet layer
• I have added an integration test which syncs a wallet for a few seconds using the 'listen' and 'watchWallet' functions available from the network and wallet layers.

Comments

[Anviking]

Nice!

[Anviking]

I fixed missing import error, but tests are failing for me

  test/integration/Cardano/Wallet/Network/HttpBridgeSpec.hs:50:13:
  1) Cardano.Wallet.Network.HttpBridge, Happy paths, get from packed epochs
       expected: Right 21600
        but got: Left (NodeUnavailable "Connection error: HttpExceptionRequest Request {\n  host                 = \"localhost\"\n  port                 = 1337\n  secure               = False\n  requestHeaders       = [(\"Accept\",\"text/plain\")]\n  path                 = \"/testnet/tip\"\n  queryString          = \"\"\n  method               = \"GET\"\n  proxy                = Nothing\n  rawBody              = False\n  redirectCount        = 10\n  responseTimeout      = ResponseTimeoutDefault\n  requestVersion       = HTTP/1.1\n}\n (ConnectionFailure Network.Socket.connect: <socket: 15>: does not exist (Connection refused))")

Ah:

thread 'main' panicked at 'start http server: Io(Os { code: 48, kind: AddrInUse, message: "Address already in use" })', src/libcore/result.rs:1009:5
note: Run with `RUST_BACKTRACE=1` for a backtrace.

Not sure if we should use different ports, or make sure they are not running at the same time.

[Anviking]

@KtorZ I added this to avoid Address already in use-problems. Should pass now, but not exactly sure what the problem was.

A failing log: https://travis-ci.org/input-output-hk/cardano-wallet/jobs/510990561

[KtorZ]

10 seconds is a rather long delay, and I wouldn't expect this to fail even without any delay between two successive runs of the spec.

[KtorZ]

🤔 ... having a look

[Anviking]

I thought that was 1s 😅

[KtorZ]

So yeah, it seems that we cancel from the async seems to wait for the delivery of the exception, but the socket isn't freed immediately by the OS. I've added a delay of 500ms which should be plenty of time already to free up the socket. I've re-run that 10x in a row without getting any issues 🤔