On 04/21/2015 12:19 AM, Justin Clift wrote:
On 20 Apr 2015, at 18:53, Jeff Darcy <jdarcy@xxxxxxxxxx> wrote:
I propose that we don't drop test units but provide an ack to patches
that have known regression failures.
IIRC maintainers have had permission to issue such overrides since a
community meeting some months ago, but such overrides have remained
rare. What should we do to ensure that currently failing Jenkins
results are checked and (if necessary) overridden in a consistent
and timely fashion, without putting all of that burden directly on
your shoulders? Some sort of "officer of the day" rotation? An
Etherpad work queue? Something else?
An Etherpad is probably a good basis for doing the listing. No
preferences personally for how it gets attended to though. :)
Another option would be to maintain a file with this list in the tests
directory. run-tests.sh can lookup this file to determine whether it
should continue or bail out.
-Vijay
_______________________________________________
Gluster-devel mailing list
Gluster-devel@xxxxxxxxxxx
http://www.gluster.org/mailman/listinfo/gluster-devel
