On Thu, May 14, 2020 at 9:59 AM David Galloway <dgallowa@xxxxxxxxxx> wrote: > > > On 5/13/20 3:20 PM, Casey Bodley wrote: > > On Wed, May 13, 2020 at 1:10 PM David Galloway <dgallowa@xxxxxxxxxx> wrote: > >> > >> > >> On 5/13/20 12:28 PM, Casey Bodley wrote: > >>> On Wed, May 13, 2020 at 11:35 AM Yuri Weinstein <yweinste@xxxxxxxxxx> wrote: > >>>> > >>>> Details of this release summarized here: > >>>> https://tracker.ceph.com/issues/45455#note-1 > >>>> > >>>> rados - approved Neha? > >>>> rgw - approved Casey? > >>> > >>> pretty much all of the failures and dead jobs are due to 'No space > >>> left on device' errors. can we find a way to address those? > >> > >> We'd have to get qty 205 larger hard drives for the smithi. > >> > >> I see we're downloading a 134GB file as part of a test. Can we stop > >> doing that? > >> > >> 2020-05-09T22:05:32.971 INFO:teuthology.orchestra.run.smithi049.stderr: > >> 92900K 100% 134G=1.0s > > > > that's downloading gradle-6.0.1-bin.zip, which is a 92M file > > Welp, that's embarrassing. > > Is it possible to logrotate more often? I think there's stuff in the > /qa dir in ceph.git that does that during tests. sorry, i remember working around these ENOSPC failures for master in https://github.com/ceph/ceph/pull/34253. i just cherry-picked that for octopus in https://github.com/ceph/ceph/pull/35067. the root cause is too much debug logging - maybe logrotate would help > > > >>> > >>>> rbd - PASSED > >>>> krbd - approved Jason, Ilya? > >>>> fs - PASSED > >>>> kcephfs - approved Patrick? > >>>> multimds - approved Patrick? > >>>> ceph-deploy - N/A > >>>> > >>>> upgrade/mimic-x (octopus) - PASSED > >>>> upgrade/nautilus-x (octopus) - PASSED > >>>> upgrade/octopus-p2p - PASSED > >>>> upgrade/client-upgrade-luminous-octopus - PASSED > >>>> upgrade/client-upgrade-mimic-octopus - PASSED > >>>> upgrade/client-upgrade-nautilus-octopus - PASSED > >>>> powercycle - in progress > >>>> ceph-ansible - approved Brad? > >>>> ceph-volume - approved Jan? (ceph-ansible bug) > >>>> > >>>> (please speak up if something is missing) > >>>> > >>>> sepia had been upgraded to this point release and performs well. > >>>> > >>>> Thx > >>>> YuriW > >>>> _______________________________________________ > >>>> Dev mailing list -- dev@xxxxxxx > >>>> To unsubscribe send an email to dev-leave@xxxxxxx > >>>> > >>> _______________________________________________ > >>> Dev mailing list -- dev@xxxxxxx > >>> To unsubscribe send an email to dev-leave@xxxxxxx > >>> > >> > > > _______________________________________________ Dev mailing list -- dev@xxxxxxx To unsubscribe send an email to dev-leave@xxxxxxx
