This is the mail archive of the
mailing list for the Cygwin project.
Re: cygcheck improvements
On Wed, Nov 02, 2005 at 05:50:28PM -0000, Dave Korn wrote:
>Christopher Faylor wrote:
>> On Wed, Nov 02, 2005 at 05:37:25PM -0000, Dave Korn wrote:
>>> Christopher Faylor wrote:
>>>> Seeing Igor's analysis of corrupt /etc/services symlinks reminded me
>>>> that I wanted to start a discussion on cygcheck improvements.
>>>> Ideally, cygcheck should report on things like /etc/services being
>>>> What other kind of common things could cygcheck be testing for?
>>> Cygwin-related entries still pending in the Replace-on-Reboot reg key
>>> might be informative.....
>> Yep. Good idea.
>> Should we run cron_diagnose.sh, too? Or, more generically, maybe we
>> should have some way for packages to register themselves with cygcheck
>> so that cygcheck would know to run them? That might be a catch-22,
> Dunno, I think that cron_diagnose is probably a bit overly-specialised to be
>run along with every single cygcheck, and in the more general case, if lots of
>packages register check-callbacks with cygcheck, we're going to end up
>(implicitly) making it dependent on a whole lot of other stuff being there and
But, if cron is installed, and cron_diagnose fails because of other things not
being available that's useful information, too. I'm not saying that cygcheck
should stop if cron_diagnose.sh doesn't run.
> Perhaps if we added it under some kind of --run-extra-tests flag that was
>off by default?
If we have to ask people to do something extra then I don't think there's
a lot of benefit. If we can ask them to use --run-extra-tests then we can
also ask them to just run cron_diagnose.sh.
Here's a scary one: How about a "cygcheck --report" option which queries
for a "from" email address and sends the report (as an attachment) to
the cygwin mailing list?
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html