Hi Sam, When running the rados suite on a given backport targetting jewel (or hammer for that matter, but it's less of an issue) I figured the smallest subset would be enough to guarantee there is no trivial regression, this is why we're routinely using X/2000. I understand this is not enough coverage to release jewel and we have the nightlies running X/14 for that. When jewel is eventually released, all of the rados suite should have been run. If you think there are too many regressions merged in the jewel branch because X/2000 did not detect them, we will increase the size of the subset. This will slow down backporting when the lab is misbehaving, but that's less time consuming than chassing a regression long after the guilty commit has been merged. Cheers [1] http://tracker.ceph.com/projects/ceph-releases/wiki/Sepia/97 -- Loïc Dachary, Artisan Logiciel Libre -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
