Hallo,
        as I spent the whole afternoon on a similar issue...  :-)

Run purge (will also remove ceph packages, I am assuming you don't care much about the existing stuff),

on all nodes mon/osd/admin remove
  rm -rf /var/lib/ceph/

on OSD nodes make sure you mount all partitions, then remove
  rm -rf /srv/node/<whatever>*/*
  chown -R ceph.ceph /srv/node/<whatever>*/

on admin node, remove, from the cluster-administration-directory,
        rm ceph.bootstrap* ceph*keyring
and only leave the old ceph.conf which you will probably susbstitute with the default one after step 1 (assuming, for example, you either spent some time playing with it and/or you want to force a specific cephfs-id).

  Good luck

                        Fulvio

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Reply via email to