Hi,
Looking to take our Octopus Ceph up to Pacific in the coming days.
All the machines (physical - osd,mon,admin,meta) are running Debian
'buster' and the setup was done originally with cephdeploy (~2016).
Previously I've been able to upgrade the core OS, keeping the ceph
packages at the same back-ported version and then when that's complete
for all the machines I've then upgraded Ceph itself.
Checking around today on download.ceph.com I cannot see any 15.2.16
packages available for Debian 'bullseye' which suggests that the upgrade
path would look more like changing Debian to 'bullseye' AND Ceph to
16.2.7 in one upgrade/dist-upgrade. With something like 400+ VMs running
live on this storage that sounds a little hairy to me.
Is it safe to use the Proxmox sources at
deb http://download.proxmox.com/debian/ceph-octopus bullseye test
and install bullseye packages with 15.2.16-pve1 over the current buster
15.2.16-1~bpo10+1 ones?
Secondarily, would anyone make a strong case for taking this opportunity
to move to cephadm management? Is there a simple path to do that from a
cluster started on Hammer? Aside from probably adding one or two OSDs
every 2/3 years this cluster is fairly static so I'm not sure it's
really worth it.
Thanks in advance.
--
All postal correspondence to:
The Positive Internet Company, 24 Ganton Street, London. W1F 7QY
*Follow us on Twitter* @posipeople
The Positive Internet Company Limited is registered in England and Wales.
Registered company number: 3673639. VAT no: 726 7072 28.
Registered office: Northside House, Mount Pleasant, Barnet, Herts, EN4 9EE.
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
