[DRE-maint] ruby-httpclient: FTBFS randomly (failing tests)

[Santiago Vila]

severity 834686 serious
thanks

Dear maintainer:

I expected a general guideline from Release Managers regarding
packages which FTBFS randomly like this one, but that will most surely
not happen for stretch.

So, the only guideline I have left is the one expressed by Julien
Cristau (one of the RMs) in Bug #844264:

"if the failure rate is low enough I think a lower severity can make
sense"

I posted a list of bugs which FTBFS more than 10% of the time and
several people agreed on -devel that they should be serious and
maintainers should ask for stretch-ignore tag in case the bug should
not be RC (most bugs are about failing tests so this should not be
needed in general).

In the latest tests I made, this package fails for me more than 85% of
the time, and it also fails (randomly) in the reproducible builds
autobuilders:

https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/ruby-httpclient.html

This may affect any user trying to build this package (for example, 
after making a change), our downstreams, and the Security Team once
that stretch is stable.

For this reason I'm raising this to serious again.

Sorry for the long explanation.

Thanks.