https://bugzilla.redhat.com/show_bug.cgi?id=1563208



--- Comment #6 from Petr Pisar <ppi...@redhat.com> ---
This makes the test to pass:

diff --git a/t/50resolver.t b/t/50resolver.t
index 7047b9d..587eb21 100644
--- a/t/50resolver.t
+++ b/t/50resolver.t
@@ -326,11 +326,13 @@ my $missinghost = "TbK4jM2M0OS.lm57DWIyu4i";
 SKIP: {
     skip "Resolver has an answer for $missinghost", 1 if gethostbyname(
$missinghost );

-    my $future = wait_for_future $resolver->getaddrinfo(
+    my $future = $resolver->getaddrinfo(
        host     => $missinghost,
        service  => "80",
        socktype => SOCK_STREAM,
+       timeout => 60,
     );
+    wait_for { $future->is_ready } timeout => 60;

     ok( $future->failure, '$future failed for missing host' );
     is( ( $future->failure )[1], "resolve", '->failure [1] gives resolve' );

Maybe one could measure the "skip" command delay and use that period instead of
the hard-coded 60 seconds. But if a resolver cached missed, and the cache
expired, the measure would be unreliable.

Nevertheless I will need this patch to apply because the current code will fail
in my weekly rebuilds.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
perl-devel mailing list -- perl-devel@lists.fedoraproject.org
To unsubscribe send an email to perl-devel-le...@lists.fedoraproject.org

Reply via email to