This is my experience with it, and may be entirely due to a misconfiguration on my part.
I stopped testing on Windows late last year when I found YACSmoke (0.64) would only run 4-6 hours before starting to send fail reports for everything. I suspect due to the PERL5LIB environment variable growing to over 20k. Since I don't want to send false failures, I stopped running the tools on Windows. I had no performance issues on the machine when it did run. Last I checked, running tests by hand from the command line worked fine on the machine. I also did not report any of the issues I had :(. I had different issues on the Pogoplug that I got mostly solved, there due to low onboard memory and having to run a tool to shove randomness into the kernel because otherwise I'd find it hung for a day if someone used ssh in their test. I stopped running that machine because of a flash-drive corruption due to power failure, and just haven't gotten back to it. I had meant to write everything up since you can have a complete computer+drive running CPAN testing (or something else) for under $50. Dana On Thu, Feb 23, 2012 at 4:11 AM, Tim Bunce <tim.bu...@pobox.com> wrote: > I was recently lamenting, on dbi-dev mailing list, the poor coverage of > cpantesters, re a recent DBI trial release: > > http://matrix.cpantesters.org/?dist=DBI+1.617_903 > > Martin Evans [CC'd] said: > >> On the Windows smokers question, I tried that a few years ago and >> found it a PITA eventually bringing my machine to a stand still. To be >> honest I found smoke testing on Linux/AIX/Solaris a PITA too but just >> not as much as Windows. Too many modules prompt even under smoke >> testing or have incorrect dependencies so my exclusion list just kept >> growing and growing. I just looked and I have 32263 smoke reports but >> you've no idea what problems were caused generating them and I >> eventually gave up - I really don't know how bingos (Chris Williams) >> etc manage to generate so many. Personally, I'd change cpan shell to >> send installation results by default but I don't really understand the >> complexities of this (even if they were anonymous). > > Have things improved? > > Tim.
