Re: Bad IO::Socket::IP test?

On 18 September 2014 04:31, bulk 88 <bul...@hotmail.com> wrote:
>>>> Date: Wed, 17 Sep 2014 22:50:19 -0400>> From: p...@greerga.m-l.org>> To: perl...@perl.org>> Subject: Bad IO::Socket::IP test?>>>>> On Wed, 17 Sep 2014, George Greer wrote:>>>> > Failures: (common-args) none>> > [default] -DDEBUGGING>> > [default] -Duseithreads>> > [default] -DDEBUGGING -Duseithreads>> > ../cpan/IO-Socket-IP/t/22timeout.t..........................FAILED>> > Non-zero exit status: 9>> > No plan found in TAP output>>>> This test has been failing on Windows ever since IO::Socket::IP was>> updated to included it. A brief investigation looks like the test is>> calling back into IO::Socket with some parameters from a prior listening>> socket and none of that appears to explain why the second (client) socket>> gets a "Bad file descriptor" error. I didn't dig further into the base>> class to find out (yet).>>>> -->> George Greer>> It fails 50% of the times I run it manually for me. It very rarely fails for> me when running harness.

I haven't seen this failure running tests manually on my own machine,
but I'm aware of the smoke failure and have already raised a ticket
for it on rt.cpan.org:

https://rt.cpan.org/Ticket/Display.html?id=98951