Bug #113429 for IO-Async: Some name resolvers can upset 50resolver.t

Tue Mar 29 09:09:08 2016 leonerd-cpan [...] leonerd.org.uk - Ticket created

Subject:

Some name resolvers can upset 50resolver.t

# Failed test '->failure [3] gives EAI_NONAME' # at t/50resolver.t line 323. # got: '-5' # expected: '-2' # Looks like you failed 1 test of 32. t/50resolver.t ............... Dubious, test returned 1 (wstat 256, 0x100) The test expects EAI_NONAME(-2) but gets EAI_NODATA(-5) While not strictly correct, the test probably ought to accept EAI_NODATA anyway, because it seems some name resolver setups cause this to happen. -- Paul Evans

Mon Apr 04 11:46:26 2016 leonerd-cpan [...] leonerd.org.uk - Correspondence added

Fixed. -- Paul Evans

Subject:

rt113429.patch

=== modified file 't/50resolver.t' --- t/50resolver.t 2015-07-31 19:02:02 +0000 +++ t/50resolver.t 2016-04-04 15:45:59 +0000 @@ -320,7 +320,10 @@ ok( $future->failure, '$future failed for missing host' ); is( ( $future->failure )[1], "resolve", '->failure [1] gives resolve' ); is( ( $future->failure )[2], "getaddrinfo", '->failure [2] gives getaddrinfo' ); - is( ( $future->failure )[3], Socket::EAI_NONAME, '->failure [3] gives EAI_NONAME' ); + + my $errno = ( $future->failure )[3]; + ok( $errno == Socket::EAI_NONAME || $errno == Socket::EAI_NODATA, '->failure [3] gives EAI_NONAME or EAI_NODATA' ) or + diag( '$errno is ' . $errno ); } my $testaddr = pack_sockaddr_in( 80, INADDR_LOOPBACK );

Mon Apr 04 11:46:26 2016 leonerd-cpan [...] leonerd.org.uk - Status changed from 'new' to 'patched'

Mon Apr 02 14:42:47 2018 leonerd-cpan [...] leonerd.org.uk - Status changed from 'patched' to 'resolved'

Mon Apr 02 14:42:47 2018 leonerd-cpan [...] leonerd.org.uk - Fixed in 0.71 added