t/81_hosts.t fails when OpenDNS is configured as DNS provider

[Christian Walde]

On Wed, 02 Oct 2013 18:17:23 +0200, Marc Lehmann <schmorp at schmorp.de>  
wrote:

> On Wed, Oct 02, 2013 at 05:55:17PM +0200, Christian Walde  
> <walde.christian at gmail.com> wrote:
>> Also do note that this is not the case of a single user. You have a  
>> number of reports on this in your cpantesters matrix:
>
> I just went through all the fails on the latest perl release, and none  
> are due to that(or even to that test).
>
> What are these failures you talk about, and what's your evidence that
> these are indeed due to opendns (or a similar service), and not due to
> some other bug?

Going through http://matrix.cpantesters.org/?dist=AnyEvent+7.05 you have 49
FAILs due to this, none of them by me even. Some excerpts:

http://www.cpantesters.org/cpan/report/c7a4f98a-25e1-11e3-884a-21c640c32a75

o2/AOL thing: http://85.183.254.9

http://www.cpantesters.org/cpan/report/48c5aa68-2322-11e3-9ae2-ef7d3e2caed2
http://www.cpantesters.org/cpan/report/434e2860-0b64-11e3-8b9a-9e6e5735fcd0

T-Online: http://62.157.140.133

http://www.cpantesters.org/cpan/report/faa106ae-11f4-11e3-9561-96fdcf634b9e

Cox ISP: http://92.242.140.2

http://www.cpantesters.org/cpan/report/39645820-0ad2-11e3-989d-cd200a9f6c7e

OpenDNS: 67.215.65.132 -> http://website-unavailable.com

A number of those i am sure are not even under the control of the user, but
mandated by their internet service provider.



> You miss the purpose of the test - the test is not, as you assume, to  
> test the host system, the test exists to check whether AnyEvent can  
> properly
> resolve on the host system.

The source code of the test

     https://metacpan.org/source/MLEHMANN/AnyEvent-7.05/t/81_hosts.t

has exactly two tests, both of them check that the result of a failed  
resolve
were overwritten by the instructions given in file indicated by the
environment variable PERL_ANYEVENT_HOSTS.

> All that your patch does is to effectively disable the whole test, which
> means it cannot even fulfill the purpose that, according to you, the
> average user expects it to have.

All my patch does is skip the test if it is conclusively shown that the
system cannot adequately communicate a failed resolve. On all other systems
it will work as before and FAIL if PERL_ANYEVENT_HOSTS is ignored.

-- 
With regards,
Christian Walde