On 06/13/2014 11:03 AM, John Griffith wrote: > On Fri, Jun 13, 2014 at 9:52 AM, Ben Nemec <[email protected]> wrote: > >> Please don't send review requests to the openstack-dev list. The >> correct procedure is outlined here: >> >> http://lists.openstack.org/pipermail/openstack-dev/2013-September/015264.html >> >> Thanks. >> >> -Ben >> >> On 06/12/2014 10:20 PM, Yuzhou (C) wrote: >>> Hi all, >>> >>> I have submit a blueprint about deferred deletion for volumes in >> cinder: https://review.openstack.org/#/c/97034/ >>> >>> The implements of deferred deletion for volumes will introduce some >> complexity, to this point, there are different options in stackers. So we >> would like to get some feedback from anyone, particularly cloud operators. >>> >>> Here, I introduce the importance of deferred deletion for volumes >> again. >>> Currently in cinder, calling the API of deleting volume means the >> volume will be deleted immediately. If the user specify a wrong volume by >> mistake, the data in the volume may be lost forever. To avoid this, I hope >> to add a deferred deletion mechanism for volumes. So for a certain amount >> of time, volumes can be restored after the user find a misuse of deleting >> volume. >>> Moreover, there are deferred deletion implements for instance in >> nova and image in glance, I think it is very common feature to protected >> important resource. >>> So I think deferred deletion for volumes is valuable, it seems to >> me that should be sufficient. >>> >>> Welcome your feedback and suggestions! >>> >>> Thanks. >>> >>> Zhou Yu >>> >>> _______________________________________________ >>> OpenStack-dev mailing list >>> [email protected] >>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev >>> >> >> >> _______________________________________________ >> OpenStack-dev mailing list >> [email protected] >> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev >> > > Ben, > > I believe Zhou Yu is actually seeking input from the broader community on > the feature itself; not the review. > > I wasn't overly enthusiastic about adding the feature and to me the > argument of "instances do it" isn't necessarily the best reason to do > something. I'm torn on whether I think it's a good feature to implement or > not given that I think there are some complexities it introduces that may > not be worthwhile. > > Anyway, I'm fairly open to hearing more input (maybe from some users in the > community?) to get some input on how valuable or needed this feature > actually is. > > Thanks, > John >
Okay, sorry for the noise then. It sounded like a review request, but if the point was to bring the discussion to a wider audience then that is obviously fine. -Ben _______________________________________________ OpenStack-dev mailing list [email protected] http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
