Upgrade Ceph cluster + radosgw from 14.2.18 to latest 15
Hi, I want to upgrade my old Ceph cluster + Radosgw from v14 to v15. But I'm not using cephadm and I'm not sure how to limit errors as much as possible during the upgrade process? Here is my upgrade steps: Firstly, upgrade from 14.2.18 to 14.2.22 (latest nautilus version) Then, upgrade it from 14.2.22 to 15.2.17 (latest octopus version) service restart order: - restart monitors (sleep ~10) - restart managers (sleep ~5) - restart metadata OSDs (sleep ~30) - restart data OSDs (sleep ~30) - restart radosgw (sleep ~20) Is there anything wrong with these steps? Does 15.2.17 is the most stable version of Octopus? Thanks.
Hi, I want to upgrade my old Ceph cluster + Radosgw from v14 to v15. But I'm not using cephadm and I'm not sure how to limit errors as much as possible during the upgrade process?
Maybe check the changelog, check upgrading notes, and continuosly monitor the mailing list? I have to do the same upgrade and eg. I need to recreate one monitor so it has the rocksdb before upgrading.
Curious, why not go to Pacific? You can upgrade up to 2 major releases in a go. The upgrade process to pacific is here: https://docs.ceph.com/en/latest/releases/pacific/#upgrading-non-cephadm-clus... The upgrade to Octopus is here: https://docs.ceph.com/en/latest/releases/octopus/#upgrading-from-mimic-or-na... Respectfully, *Wes Dillingham* wes@wesdillingham.com LinkedIn <http://www.linkedin.com/in/wesleydillingham> On Tue, May 9, 2023 at 3:25 AM Marc <Marc@f1-outsourcing.eu> wrote:
Hi, I want to upgrade my old Ceph cluster + Radosgw from v14 to v15. But I'm not using cephadm and I'm not sure how to limit errors as much as possible during the upgrade process?
Maybe check the changelog, check upgrading notes, and continuosly monitor the mailing list? I have to do the same upgrade and eg. I need to recreate one monitor so it has the rocksdb before upgrading.
_______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
Because pacific has performance issues
Curious, why not go to Pacific? You can upgrade up to 2 major releases in a go.
The upgrade process to pacific is here: https://docs.ceph.com/en/latest/releases/pacific/#upgrading-non-cephadm- clusters The upgrade to Octopus is here: https://docs.ceph.com/en/latest/releases/octopus/#upgrading-from-mimic- or-nautilus
Hi, I want to upgrade my old Ceph cluster + Radosgw from v14 to
v15. But
I'm not using cephadm and I'm not sure how to limit errors as much as possible during the upgrade process?
Maybe check the changelog, check upgrading notes, and continuosly monitor the mailing list? I have to do the same upgrade and eg. I need to recreate one monitor so it has the rocksdb before upgrading.
Yes, the documents show an example of upgrading from Nautilus to Pacific. But I'm not really 100% trusting the Ceph documents, and I'm also afraid of what if Nautilus is not compatible with Pacific in some operations of monitor or osd =)
why are you still not on 14.2.22?
Yes, the documents show an example of upgrading from Nautilus to Pacific. But I'm not really 100% trusting the Ceph documents, and I'm also afraid of what if Nautilus is not compatible with Pacific in some operations of monitor or osd =)
I just want the latest minor version before upgrading to the next major version :) This practice isn't recommended elsewhere, but I want to make sure and limit errors as much as possible.
I have upgraded dozens of clusters 14 -> 16 using the methods described in the docs, and when followed precisely no issues have arisen. I would suggest moving to a release that is receiving backports still (pacific or quincy). The important aspects are only doing one system at a time. In the case of monitors ensuring it rejoins quorum after restarting on new version before proceeding to next mon. In the case of OSDs waiting for all PGs to be active+clean* before proceeding to the next host. Respectfully, *Wes Dillingham* wes@wesdillingham.com LinkedIn <http://www.linkedin.com/in/wesleydillingham> On Mon, May 15, 2023 at 3:46 AM Marc <Marc@f1-outsourcing.eu> wrote:
why are you still not on 14.2.22?
Yes, the documents show an example of upgrading from Nautilus to Pacific. But I'm not really 100% trusting the Ceph documents, and I'm also afraid of what if Nautilus is not compatible with Pacific in some operations of monitor or osd =)
_______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
participants (3)
-
Marc
-
viplanghe6＠gmail.com
-
Wesley Dillingham