Ceph Nautilus : Add or Remove Monitor Nodes2021/08/26 |
This is how to add or remove Monitor Daemons from exisiting Cluster.
| +--------------------+ | +----------------------+ | [dlp.srv.world] |10.0.0.30 | 10.0.0.31| [www.srv.world] | | Ceph Client +-----------+-----------+ RADOSGW | | | | | | +--------------------+ | +----------------------+ +----------------------------+----------------------------+ | | | |10.0.0.51 |10.0.0.52 |10.0.0.53 +-----------+-----------+ +-----------+-----------+ +-----------+-----------+ | [node01.srv.world] | | [node02.srv.world] | | [node03.srv.world] | | Object Storage +----+ Object Storage +----+ Object Storage | | Monitor Daemon | | | | | | Manager Daemon | | | | | +-----------------------+ +-----------------------+ +-----------------------+ |
[1] | For example, Add a [node04] node for Monitor Daemon on Admin Node. |
# transfer public key root@node01:~# ssh-copy-id node04 # if Firewalld is running, allow service root@node01:~# ssh node04 "apt update; apt -y install ceph"
# install required packages root@node01:~# FSID=$(grep "^fsid" /etc/ceph/ceph.conf | awk {'print $NF'}) root@node01:~# NODENAME="node04" root@node01:~# NODEIP="10.0.0.54" root@node01:~# monmaptool --add $NODENAME $NODEIP --fsid $FSID /etc/ceph/monmap monmaptool: monmap file /etc/ceph/monmap monmaptool: set fsid to 92749530-d9af-4226-bfe0-ccc79a689a66 monmaptool: writing epoch 0 to /etc/ceph/monmap (2 monitors) # configure monitor map root@node01:~# scp /etc/ceph/ceph.conf node04:/etc/ceph/ceph.conf root@node01:~# scp /etc/ceph/ceph.mon.keyring node04:/etc/ceph root@node01:~# scp /etc/ceph/monmap node04:/etc/ceph root@node01:~# ssh node04 "ceph-mon --cluster ceph --mkfs -i node04 --monmap /etc/ceph/monmap --keyring /etc/ceph/ceph.mon.keyring" root@node01:~# ssh node04 "chown -R ceph. /etc/ceph /var/lib/ceph/mon" root@node01:~# ssh node04 "ceph auth get mon. -o /etc/ceph/ceph.mon.keyring" root@node01:~# ssh node04 "systemctl enable --now ceph-mon@node04" root@node01:~# ssh node04 "ceph mon enable-msgr2"
ceph -s cluster: id: 92749530-d9af-4226-bfe0-ccc79a689a66 health: HEALTH_OK services: mon: 2 daemons, quorum node01,node04 (age 7s) mgr: node01(active, since 42m) mds: cephfs:1 {0=node01=up:active} osd: 3 osds: 3 up (since 9m), 3 in (since 10m) rgw: 1 daemon active (www) task status: scrub status: mds.node01: idle data: pools: 8 pools, 248 pgs objects: 237 objects, 6.7 MiB usage: 3.0 GiB used, 237 GiB / 240 GiB avail pgs: 248 active+clean |
[2] | To remove a Monitor Daemon from existing Cluster, run commands like follows. For example, Remove [node04] node. |
root@node01:~# ceph -s cluster: id: 92749530-d9af-4226-bfe0-ccc79a689a66 health: HEALTH_OK services: mon: 2 daemons, quorum node01,node04 (age 7s) mgr: node01(active, since 42m) mds: cephfs:1 {0=node01=up:active} osd: 3 osds: 3 up (since 9m), 3 in (since 10m) rgw: 1 daemon active (www) task status: scrub status: mds.node01: idle data: pools: 8 pools, 248 pgs objects: 237 objects, 6.7 MiB usage: 3.0 GiB used, 237 GiB / 240 GiB avail pgs: 248 active+clean # remove Monitor Daemon root@node01:~# ceph mon remove node04 removing mon.node04 at [v2:10.0.0.54:3300/0,v1:10.0.0.54:6789/0], there will be 1 monitors # disable monitor daemon root@node01:~# ssh node04 "systemctl disable --now ceph-mon@node04.service" ceph -s cluster: id: 92749530-d9af-4226-bfe0-ccc79a689a66 health: HEALTH_OK services: mon: 1 daemons, quorum node01 (age 25s) mgr: node01(active, since 43m) mds: cephfs:1 {0=node01=up:active} osd: 3 osds: 3 up (since 10m), 3 in (since 11m) rgw: 1 daemon active (www) task status: scrub status: mds.node01: idle data: pools: 8 pools, 248 pgs objects: 237 objects, 6.7 MiB usage: 3.0 GiB used, 237 GiB / 240 GiB avail pgs: 248 active+clean |
Sponsored Link |