> -----Original Message----- > From: Steve Hay [mailto:[EMAIL PROTECTED] > Sent: 20 October 2005 12:20 > To: Steve Hay > Cc: [EMAIL PROTECTED]; [email protected]; Paul Marquess > Subject: Re: Smoke [5.9.3] 25808 FAIL(F) MSWin32 WinXP/.Net SP2 (x86/2 > cpu) > > Steve Hay wrote: > > Automated smoke report for 5.9.3 patch 25808 > > Mugwump.uk.radan.com: Intel(R) Pentium(R) 4 CPU 3.40GHz(~3391 MHz) > (x86/2 cpu) > > on MSWin32 - WinXP/.Net SP2 > > using cl version 12.00.8804 > > smoketime 10 hours 57 minutes (average 16 minutes 26 seconds) > > > > Summary: FAIL(F) > [...] > > ../lib/Archive/Tar/t/02_methods.t.......FAILED 16-23 26 > > I think these failures could be due to the new Compress-Zlib > distribution in bleadperl. > > lib/Archive/Tar.pm was changed in 25312 (upgrading to 1.26) and hasn't > been touched since. > > I have a bleadperl smoke report at 25698 in which all tests passed OK. > > Compress-Zlib then went through a period of crashing in blead on Win32. > By the time it was fixed (change 25758), the Archive-Tar tests had > started failing (e.g. the failure is in a smoke report at 25759).
I assume that "Inconsistent test results (between TEST and harness)" means you've tested it both via "make test" and by running t/TEST directly? If so, how is t/TEST run? sorry for the delay in following up on this. Paul
