Shulou(Shulou.com)06/03 Report--

1. Delete osd

Delete the created osd.

Osd with data and logs on the same disk

Kick osd.0 out of the cluster and execute ceph osd out 0

Stop this osd process and execute systemctl stop ceph-osd@0

Then execute: ceph osd crush remove osd.0, when osd.0 is no longer in osd tree

Execute ceph auth del osd.0 and ceph osd rm 0, and delete successfully, but the original data and log directories are still there, that is, the data is still there.

At this point we will / dev/sdd disk umount, and then erase the disk so that the data will be completely deleted, execute umount / dev/sdd, and then execute ceph-disk zap / dev/sdd

However, it was found that umount could not be used and the folder could not be deleted.

Method

Create osd

Ceph-deploy osd create-data / dev/sdd node1

Because there is an original lvm record, it cannot be recreated. It is suggested that the original lvm has a record.

Record the vg serial number

Ceph-volume lvm list

Delete vg record

Vgremove ceph-7216ab35-9637-4930-9537-afe9f1525efa

That's when sdd came out.

Create an osd in the secondary execution

Ceph-deploy osd create-data / dev/sdd node1

Osd.0 in joins the cluster

Ceph osd in osd.0

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.