On Sat, Oct 30, 2021 at 11:53:46AM +0300, Adrian Bunk wrote: > Source: ruby-eventmachine > Severity: serious > Tags: ftbfs > > FTBFS on IPV6-only buildds: > > https://buildd.debian.org/status/fetch.php?pkg=ruby-eventmachine&arch=amd64&ver=1.3~pre20201020-b50c135-3&stamp=1633500476&raw=0 > https://buildd.debian.org/status/fetch.php?pkg=ruby-eventmachine&arch=i386&ver=1.3~pre20201020-b50c135-4%2Bb1&stamp=1635578996&raw=0 > > FTBFS due to attempted DNS: > > https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/ruby-eventmachine.html > > ... > =============================================================================== > Failure: test_nameserver(TestResolver): > <nil> was expected to be kind_of? > <String> but was > <NilClass>. > /build/1st/ruby-eventmachine-1.3~pre20201020-b50c135/tests/test_resolver.rb:19:in > `test_nameserver' > 16: end > 17: > 18: def test_nameserver > => 19: assert_kind_of(String, EM::DNS::Resolver.nameserver) > 20: end > 21: > 22: def test_nameservers > =============================================================================== > ...
The ipv6-only thing is already worked around in 1.3~pre20201020-b50c135-4 by skipping the tests in those setups. Now, there is some type of weird issue with the test suite. It seems to fail randomly on some circumstances, which I could not pinpoint yet. e.g. I cannot reproduce the failures on a i386 chroot locally.
signature.asc
Description: PGP signature
