Thanks all for the replies :) Indeed it looks like docker hub has been deprecated, and we are now on quay.io although I would say the communications were a bit lacking, and it's been implemented really quick leaving things like Cephadmin on the Octopus branch broken (pending backport) for use with 15.2.14 I did post on the tracker as well, and did some digging - linking here for completeness as it has PR# in it for fixing the now "broken" cephadmin : https://tracker.ceph.com/issues/52287 To give some completeness to the thread, we picked containers as it was "easier" with our host standard build, to blast Ceph out into the farm with the minimum technical intervention - we are also happy supporting Docker as we use it natively for allot of other things so it made sense at the time :) The docker pull issue on bandwidth was a pain, but compared to the slow down we have seen on quay.io when we eventually applied 15.2.14 to the cluster, well... it was actually slower, allot slower (docker hub was about 1m-2m) (quay.io was so slow, I thought the upgrade had hung ! it was stuck on the node being upgraded for a good 5-8m pulling image) again we don't mind as the ceph container upgrade is a thing a beauty to watch spreading over the cluster :) [Note : the bug with Managers being on different architectures X86 and Arch64 is back in 15.2.14, so had to fiddle moving them all over onto same architecture to make the upgrade continue...] I guess, by moving onto Redhats container store, far more interesting telemetry might be available to the teams as well... Anyway, thanks for the responses :) Jamie _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |