How to deploy ceph-deploy
This article shows you how to deploy ceph-deploy, the content is concise and easy to understand, can definitely brighten your eyes, through the detailed introduction of this article, I hope you can get something.
Environment description: ceph2 and ceph3 servers (where ceph2 has installed ceph-deploy)
One S3500 SSD and two Hitachi 1T HDD disks on each server
Integration of front and back end and management network: 128.128.128.9 and 128.128.10, respectively
1. Add: in / etc/hosts:
128.128.128.9 ceph2128.128.128.10 ceph3
2. Pass
Ssh-keygen and ssh-copy-id
Make the nodes secret-free
3. Create a new directory in ceph2 and keep the configuration file
Mkdir / root/myclustercd / root/mycluster
4. Install ceph
Ceph-deploy install ceph2 ceph3
5. Install mon
Ceph-deploy new ceph2
6. Modify ceph.conf configuration file
Public_network = 128.128.128.0/24cluster_network = 128.128.128.0/24enable experimental unrecoverable data corrupting features = bluestore rocksdb debug_white_box_testing_ec_overwritesbluestore block db size = 10737418240 # 10Gbluestore block wal size = 10737418240 # 10Gosd objectstore = bluestoremon_allow_pool_delete = truerbd_cache = false [osd] bluestore = true
7. Initialize mon
Ceph-deploy mon create-initial
If you need to add mon
Ceph-deploy add mon ceph3
8. Format the partition
Ceph-deploy zap {ceph-node}: {dest-disk} for example: ceph-deploy zap ceph2:/dev/sdb
9. Copy ceph.bootstrap-osd.keyring to / var/lib/ceph/bootstrap-osd/ and command ceph.keyring
Cp / home/mycluster/ceph.bootstrap-osd.keyring / var/lib/ceph/bootstrap-osd/ceph.keyring or scp / home/mycluster/ceph.bootstrap-osd.keyring ceph3:/var/lib/ceph/bootstrap-osd/ceph.keyring
10. Add osd
Ceph-disk prepare-bluestore-block.db / dev/sdb-block.wal / dev/sdb / dev/sdc where: / dev/sdb is SSD disk, / dev/sdc is HDD disk. The above is how to deploy ceph-deploy. Have you learned any knowledge or skills? If you want to learn more skills or enrich your knowledge reserve, you are welcome to follow the industry information channel.
