all of the abuse related repos (and others actually) need mongo, so there's a make command called start-mongo that you'll probably need to run before running the tests. Is it possible that was the issue the tests appeared broken and time out

Originally posted by @peterjan in #42 (comment)

This times out for me at 30s, so we should increase the timeout.

They are also failing when I do increase the timeout. @peterjan are they supposed to pass?

Originally posted by @ro-tex in #42 (comment)

This is pretty futile. Because we are not using a waitgroup in this function. So nothing is waiting for the thread to finish which in turn means it will be killed as soon as main exits.

Originally posted by @ChrisSchinnerl in #19 (comment)

I think we should refactor this function and make it a lot cleaner.
I don't mind doing that I just didn't want to make too many changes to this repo.

Originally posted by @peterjan in #7 (comment)

Client is a generic name that would indicate that this is some sort of client for talking to blocker, rather than an external service like skyd, especially having in mind that it will be used as api.Client. Calling it SkydClient or PortalClient would be more informative.

Alternatively, we can move most (all?) of these types to a new package called skyd and then their names will make perfect sense.

We are even using it as skydClient api.Client.

Originally posted by @ro-tex in #27 (comment)

Overview

Design or Proposal

Might warrant a generic .post method like we are already doing for get requests.

Originally posted by @ChrisSchinnerl in #19 (comment)