Absolutely. I’ve migrated clusters, eg. from Ubuntu to RHEL, and from Ubuntu Trusty to Bionic, in a rolling fashion. I strongly suggest arranging for the Ceph version to be *exactly* the same on both OS releases, so that you’re only changing one variable at a time. It appears that there is a way to upgrade CentOS 7 to 8 without re-imaging the boot drive. If you go to Ubuntu, you’ll likely have to re-image. Either way, ensure that your OSD drives are *not* wiped by the process. If you proceed as you describe, adopting first, I *think* you can bring back the OSDs with `ceph-volume lvm activate`. I’m less sure about how to manage the mon, mgr, and iscsi-gw services; perhaps someone else can chime in. I do suggest that you test out the plan on a lab/dev cluster first, and that you give the cluster some time, say, a week, between Ceph and OS upgrades, just out of an abundance of cautionoia. — aad > > Hi, > > I had Ceph clusters running with Nautilus with CentOS 7. But I would like > to enjoy new features from later versions. (e.g. enhanced iSCSI > performance.) > > Since Ceph Octopus no longer fully supports CentOS 7. I am targeted to > migrate the Ceph cluster from CentOS 7 to CentOS 8 or Ubuntu. > I would like to know if Ceph is capable of running under different OS > families. > > Are there any practices for OS upgrade/migration that can be found from the > official site? > > My drafted plan is: > 1. [CentOS 7] Adopt the cluster by cephadm via upgrading it from Nautilus > to Octopus. > 2. Reinstall the nodes one by one (with new OS) > > Regs, > Icy > _______________________________________________ > ceph-users mailing list -- ceph-users@xxxxxxx > To unsubscribe send an email to ceph-users-leave@xxxxxxx _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |