Even though the test doesn't support windows, I had to make it compilable for windows. I've manually tested it on my windows PC and it is properly killing daemon.

@onee-only thank you for looking into this. Please note that plumbing/transport is being refactored and the latest version lives in the v6-transport branch. Please rebase and target that branch instead.

@pjbgf Thanks for telling me that I wasn't targeting the appropriate branch.

I've rebased it to v6-transport. However, since the tests are properly executed, tests are failing with latest changes on this branch.
I tried to dig into it, but I think this is out of my hand.

FYI, ReceivePackSuite in transport/file stalls when it executes TestSendPackAddDeleteReference.
I suspect it is related to gooutines created in (c *command) Start(), which is in transport/file/client.go

@pjbgf It seems that this PR has conflicting changes with #1496.
I believe at least 25e63f8 is necessary for v6-transport as this as it exposes hidden failing tests.

@onee-only apologies for the wasted effort here. @aymanbagabas have been working on several changes across transport, so making changes to tests that touch any of that is a bit problematic. Perhaps it is easier to just revisit this once #1496 is merged?

I found out that #1496 introduces partial fix for #1445 by changing suites on git package.
Since 25e63f8 is mainly about "inheriting the test suite properly", I think we can just abondon this commit and make #1496 fix the test suites in transport/xxx package.
It'll be more convinient for @aymanbagabas to fix the failing tests in the same PR.

About #1218, I didn't test it but fix for it also seems to be introduced in #1496.

So I think I'll close this PR and make new one with small changes when #1496 is merged.
@pjbgf What are your thoughts?

@onee-only that sounds good, thanks for that. 🙇

Alright. closing the PR.