Inline tests fail with "HARNESS_OPTIONS=j9"

[David Oswald]

This might seem obvious, but I stumble on it from time to time. Inline's
testing fails if the environment variable HARNESS_OPTIONS has "jn" set (n
being number of simultaneous processes permitted).

There seem to be fewer and fewer modules for which this is a problem. I
think that many compilers are not happy with being run in parallel, so
probably the best way to handle it is to have our Makefile.PL set 'make
test' to temporarily unset HARNESS_OPTIONS. Does anyone have any idea how
to do that?

Dave

[Sisyphus]

----- Original Message -----
From: "David Oswald" <daoswald-***@public.gmane.org>
To: "inline" <inline-***@public.gmane.org>
Sent: Sunday, November 18, 2012 6:07 AM
Subject: Inline tests fail with "HARNESS_OPTIONS=j9"

Post by David Oswald
This might seem obvious, but I stumble on it from time to time. Inline's
testing fails if the environment variable HARNESS_OPTIONS has "jn" set (n
being number of simultaneous processes permitted).
There seem to be fewer and fewer modules for which this is a problem. I
think that many compilers are not happy with being run in parallel, so
probably the best way to handle it is to have our Makefile.PL set 'make
test' to temporarily unset HARNESS_OPTIONS. Does anyone have any idea how
to do that?

I don't know how to do this with 'make test', but the test scripts
themselves should be able to query and alter $ENV{HARNESS_OPTIONS}.
I think that would achieve the desired result, so long as any such fiddling
is done in a BEGIN{} block at the very start of the script.

Cheers,
Rob