Ok so here's the clarification: First of all, by "dropping NFSv3 support (by default)" I didn't mean that NFSv3 wouldn't be supported anymore, but that it would be disabled by default and would need to be explicitly enabled in the service specs. Also, this is not a future change, it already happened. v20.2.3 ships PR #70675, so upgrading to v20.2.3 silently disables NFSv3. The related PRs were merged shortly before the tag was cut and are missing from the v20.2.3 release notes and changelog, so there is currently no warning anywhere. We should fix that. Users who need NFSv3 must now explicitly enable it by adding 'enable_nfsv3: true' to the NFS service spec. Since older MGRs reject this field, it can only be applied once MGRs run v20.2.3. The cleanest path is a staggered upgrade to avoid redeploying the NFS service twice: 1. ceph orch upgrade start --image <v20.2.3> --daemon-types mgr 2. add 'enable_nfsv3: true' to the NFS service spec and re-apply it 3. run the upgrade as usual (already upgraded MGRs will be skipped) Users who have already upgraded should simply update the spec and re-apply it to get NFSv3 back. @Wodel Ultimately, upgrading to v20.2.3 should address your original concern. Best regards, Frédéric. -- Frédéric Nass Ceph Ambassador France | Senior Ceph Engineer @ CLYSO Le sam. 8 août 2026, 02:36, Frédéric Nass via ceph-users <ceph-users@ceph.io> a écrit :

Hi Robert,

Tentacle should soon be dropping NFSv3 support (by default) as per https://github.com/ceph/ceph/pull/70675. My understanding so far is that users will have to add 'enable_nfsv3: true' in the NFS service specs to keep using NFSv3. I hope that can soon be clarified in the PR and Tentacle documentation.

Cheers, Frédéric

-- Frédéric Nass

Ceph Ambassador France | Senior Ceph Engineer @ CLYSO

Frédéric Nass

Senior Ceph Engineer

Ceph Ambassador, France

+49 89 215252-751 <https://call.ctrlq.org/+49%2089%20215252-751>

frederic.nass@clyso.com

www.clyso.com

Hohenzollernstr. 27, 80801 Munich <https://www.google.com/maps/search/Hohenzollernstr.+27,+80801+Munich?entry=gmail&source=g>

Utting a. A. | HR: Augsburg | HRB: 25866 | USt. ID-Nr.: DE2754306

Le ven. 7 août 2026, 18:07, Robert Sander via ceph-users < ceph-users@ceph.io> a écrit :

...

Hi,

Am 07.08.26 um 10:59 AM schrieb Frédéric Nass via ceph-users:

...

Or, if that's too much of a maintenance burden and you're worried you'll forget that step, just install rpcbind on the NFS nodes and leave NFSv3 enabled.

Or maybe NFSv3 is made an option for the NFS service spec with the hint that then rpcbind has to be installed.

Or the orchestrator installs rpcbind when the NFS service should offer NFSv3.

Regards -- Robert Sander Linux Consultant

Heinlein Consulting GmbH Schwedter Str. 8/9b, 10119 Berlin

https://www.heinlein-support.de

Tel: +49 30 405051 - 0 Fax: +49 30 405051 - 19

Amtsgericht Berlin-Charlottenburg - HRB 220009 B Geschäftsführer: Peer Heinlein - Sitz: Berlin _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io

_______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io

Frédéric Nass Senior Ceph Engineer Ceph Ambassador, France +49 89 215252-751 <https://call.ctrlq.org/+49%2089%20215252-751> frederic.nass@clyso.com www.clyso.com Hohenzollernstr. 27, 80801 Munich Utting a. A. | HR: Augsburg | HRB: 25866 | USt. ID-Nr.: DE2754306