We run
ceph osd crush reweight osd.{odd-num} 0
before
ceph osd out {osd-num}
to avoid double cluster rebalancing after
ceph osd crush remove osd. {osd-num}
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users
Hi all
On a Ceph cluster deployed wit ceph-deploy the instructions in the doku to
remove a osd seem outdated to us.
We did the following steps:
ceph osd out {osd-num}
sudo /etc/init.d/ceph stop osd.{osd-num} -> WON'T work as there is no osd.disk
declaration in ceph.conf hast to be replaced b
