Re: svn commit: r15659 - trunk/subversion/tests/libsvn_fs

On Fri, 11 Aug 2005, C. Michael Pilato wrote:

> Jani Averbach <jaa@jaa.iki.fi> writes:
> > And now of course BDB passes unexpectedly. So I see few options:
> >
> > 1) Fix the actual problem, and stop tweaking the test suite.
>
> Funny. I only see one option in my mail buffer...
>
Implying that the failures will continue on the breakage list. Doyou mean
that everone who committed should have to examine every failure mail to
check if it is just that FSFS problem? Isn't that a waste of peoples'
time?

Normally, when we find a bug, we ad an XFail (at best) and file an issue.
What makes this bug different? Is it extremely critical?

I don't see "well, go fix that damn FSFS bug" as a reasonable answer,
since I don't have the knowledge to do so.

I have no problem at all with test that fail for some days becuase people
might not be able to fix things in a hurry (or might not even notice over
the weekend or something), but deliberately letting tests fail for weeks
doesn't seem very helpful.

If the tone of this mail sounds frustrated, well... I'm just a *little*
frustrated:-)

Best,
//Peter

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Fri Aug 12 12:07:18 2005