Find a way to avert the CI failure No output has been received in the last 10 minutes

[DavidGregory084]

CI builds occasionally fail due to the message:

No output has been received in the last 10 minutes, this potentially indicates a stalled build or something wrong with the build itself.

This is off-putting for newcomers as it's not immediately obvious that there might be nothing wrong with their PR.

Could we avert this by using the travis_wait command or is there some other way that we could speed up builds so they don't encounter this timeout?

[ghost]

This is caused by a know bug that has been fixed in scalacheck, but not released. Might be best to wait on this for now.

[DavidGregory084]

Ah, that makes sense. Shall we just close this, or wait for typelevel/scalacheck#195?

[paulp]

Looking at typelevel/scalacheck#195 I see this has been going on for six months. That's about 5.95 months too long when it comes to nondeterministic CI failures. They are huge time wasters.

[DavidGregory084]

@paulp actually 1.13.0 is released now and it looks like cats is building against 1.12.5, maybe just needs a version bump?

[ceedubs]

@DavidGregory084 I suspect we are going to run into typelevel/scalacheck#218 if we do that.

[DavidGregory084]

😢

[paulp]

It doesn't even compile against 1.13.0. A little excerpt.

[error] /s/scala/cats/laws/src/main/scala/cats/laws/discipline/TraverseTests.scala:37: could not find implicit value for parameter a2: org.scalacheck.Arbitrary[A => M]
[error]         "traverse derive foldMap" -> forAll(laws.foldMapDerived[A, M] _)
[error]                                            ^
[error] /s/scala/cats/laws/src/main/scala/cats/laws/discipline/TraverseTests.scala:36: could not find implicit value for parameter a2: org.scalacheck.Arbitrary[A => X[B]]
[error]         "traverse parallel composition" -> forAll(laws.traverseParallelComposition[A, B, X, Y] _),
[error]                                                  ^
[error] 50 errors found

[ceedubs]

@paulp I haven't really looked yet, so I could be totally wrong, but my guess is that those compile errors are in places where we've assumed that if we have an Arbitrary[A] we have an Arbitrary[X => A]. That was true with previous versions of Scalacheck, but only because it only produced arbitrary functions with a constant result value. I'm guessing we'll have to pass along those arbitrary function requirements (because we don't already have enough implicit parameters in there...).

[xuwei-k]

You should define org.scalacheck.Cogen[X] and Arbitrary[A] if you need Arbitrary[X => A].

• Support non-constant arbitrary functions scalacheck#171
• https://github.com/rickynils/scalacheck/blob/1.13.0/src/main/scala/org/scalacheck/ArbitraryArities.scala#L12

[ceedubs]

Thanks, @xuwei-k.

[adelbertc]

An attempt: #942

[ceedubs]

I haven't seen this problem after some scala.js and scalacheck config changes that were introduced in recent months. I'm going to go ahead and close this out. Feel free to reopen it if it becomes a problem again.