On Tuesday, August 12, 2003, at 12:15PM, John Peacock <[EMAIL PROTECTED]> wrote:
>Bad news on the VMS front with the latest snapshot. :~( Thanks for the additional testing, and consider how picky we've gotten that 99.39% successful is considered bad news. I get all tests successful on OpenVMS Alpha 7.2-1 and 7.3-1. I get two test failures on OpenVMS Alpha 7.1, a six-year-old release that needs various patches applied (to the system, not Perl). > Attached is the output >of 'perl harness' so you can see all the damage. Do you get the same failures from "MMK test" that you get from the harness? I haven't tested the harness in some time and wouldn't be surprised if some additional porting might be required. >I don't know how much time I'm >going to have to help track this down, but anything else I can specify about our >system, please let me know. I am particularly mystified about all the of the >Devel::PPPort related errors, since if I go into that directory, it tests fine. > >I edited and used the PERL_SETUP.COM script so that there would be no question >which libraries were being access. I've also attached the output of 'perl -"V"' >for the tested version. I don't think PERL_SETUP.COM is going to have any effect on running the test suite (though it does get used by MMK install). The value of the PERL_ROOT logical name will determine which Perl you're using when you type "perl harness".
