Sure, here's one where I can't reproduce the issue myself, so I don't know what the causal factor is: https://www.cpantesters.org/cpan/report/7921c70e-73ee-11ef-8c10-2b036e8775ea
The default error messages are not helpful, but when AUTOMATED_TESTING=1 is set I dump lots of data which will give me a clue how to reproduce the problem. On Tue, Sep 17, 2024 at 1:10 PM James E Keenan <jkee...@pobox.com> wrote: > On 9/17/24 15:58, Karen Etheridge wrote: > > Please please please, could everyone running automated smokers, set > > AUTOMATED_TESTING=1 in the environment? As per the Lancaster Consensus, > > this is a signal to the installer that no human is behind the > > installation, and most likely a test report will be generated, so it > > allows the installation to enable different or more verbose diagnostics. > > > > https://xdg.me/the-annotated-lancaster-consensus/ > > <https://xdg.me/the-annotated-lancaster-consensus/> > > > > I have received some perplexing failure reports lately which would be > > much easier to debug if AUTOMATED_TESTING=1 was set, as I provide a lot > > of data dumps in the test output when this value is set. As it is I am > > unable to determine what particular combination of environment in the > > smoker is resulting in these failures, so I am unable to fix them. > > > > thank you, > > > > a CPAN author > > -ether > > > > Could you provide URLs to a couple of such reports that would have been > more helpful if that envvar had been set? > > I ask because I have never consciously set that envvar and I'd like to > know if the manually-triggered reports I do generate are adequate. > > Thanks. > jimk >
