- 0.准备集群
- 1.环境预检
- 2.搭建集群
- 3.Dashboard配置
- 4.pool创建及使用
- 参考资料
0.准备集群
vm: 3台 CentOS-7.2-x64(3.10.0-693.5.2.el7.x86_64) 1核, 1G Mem, 20G SSD x 2
subnet: 10.0.4.0/26
| 主机 | IP | 功能 |
|---|---|---|
| node1 | 10.0.4.14(内网), 192.168.20.52 100Mbps (公网) | ceph-deploy, mon, mgr, osd |
| node2 | 10.0.4.6 (内网), 192.168.20.59 100Mbps (公网) | mon, osd |
| node3 | 10.0.4.15(内网), 192.168.20.58 100Mbps (公网) | mon, osd |
- 修改主机名(所有节点)，参考 如何在CentOS 7上修改主机名
[root@node{1,2,3} ~]# hostnamectl set-hostname node1
[root@node{1,2,3} ~]# hostnamectl --pretty
[root@node{1,2,3} ~]# hostnamectl --static
node1
[root@node{1,2,3} ~]# hostnamectl --transient
node1
[root@node{1,2,3} ~]# cat /etc/hosts
...
127.0.0.1 node1
- 创建用户，参考 CREATE A CEPH DEPLOY USER
[root@node{1,2,3} ~]# sudo useradd -d /home/search -m search # search 用户
[root@node{1,2,3} ~]# sudo passwd search
[root@node{1,2,3} ~]# echo "search ALL = (root) NOPASSWD:ALL" | sudo tee /etc/sudoers.d/search
search ALL = (root) NOPASSWD:ALL
[root@node{1,2,3} ~]# sudo chmod 0440 /etc/sudoers.d/search
- 实现SSH无密登陆，参考 ENABLE PASSWORD-LESS SSH
[root@node1 ~]# su search
[search@node1 ~]$ cat ~/.ssh/config
#~/.ssh/config
Host node2
Hostname 10.0.4.6
User search
Host node3
Hostname 10.0.4.15
User search
[search@node1 ~]$ chmod 0600 ~/.ssh/config
[search@node1 ~]$ ssh-copy-id search@node2
[search@node1 ~]$ ssh-copy-id search@node3
1.环境预检
TODO: 防火墙设置
- 配置ntp服务，参考 CEPH NODE SETUP
[search@node{1,2,3} ~]$ sudo yum install ntp ntpdate -y
[search@node{1,2,3} ~]$ sudo ntpdate pool.ntp.org
[search@node{1,2,3} ~]$ sudo systemctl enable ntpd.service
[search@node{1,2,3} ~]$ sudo systemctl enable ntpdate.service
[search@node{1,2,3} ~]$ sudo systemctl start ntpd.service
- PRIORITIES/PREFERENCES
[search@node{1,2,3} ~]$ sudo yum install yum-plugin-priorities -y
- 配置源，参考 RPM PACKAGES, INSTALLING WITH RPM
[search@node{1,2,3} ~]$ cat /etc/yum.repos.d/ceph.repo
[ceph]
name=Ceph packages for $basearch
# luminous, centos7
baseurl=https://download.ceph.com/rpm-luminous/el7/$basearch
enabled=1
priority=2
gpgcheck=1
gpgkey=https://download.ceph.com/keys/release.asc
[ceph-noarch]
name=Ceph noarch packages
baseurl=https://download.ceph.com/rpm-luminous/el7/noarch
enabled=1
priority=2
gpgcheck=1
gpgkey=https://download.ceph.com/keys/release.asc
[ceph-source]
name=Ceph source packages
baseurl=https://download.ceph.com/rpm-luminous/el7/SRPMS
enabled=0
priority=2
gpgcheck=1
gpgkey=https://download.ceph.com/keys/release.asc
- 更新源，安装依赖
[search@node{1,2,3} ~]$ sudo yum makecache
[search@node{1,2,3} ~]$ sudo yum update -y
[search@node{1,2,3} ~]$ sudo yum install snappy leveldb gdisk python-argparse gperftools-libs -y
- 安装ceph-deploy(node1)，参考 INSTALL CEPH DEPLOY
[search@node1 ~]$ sudo yum install -y ceph-deploy
[search@node1 ~]$ ceph-deploy --version
2.0.0
2.搭建集群
- 创建集群操作目录
[search@node1 ~]$ mkdir my-cluster
[search@node1 ~]$ cd my-cluster
- 创建集群，参考 CREATE A CLUSTER
[search@node1 my-cluster]$ cat /etc/hosts # 配置node1 /etc/hosts
...
10.0.4.14 node1
10.0.4.6 node2
10.0.4.15 node3
[search@node1 my-cluster]$ ceph-deploy new node1 node2 node3
[search@node1 my-cluster]$ ls
ceph.conf ceph-deploy-ceph.log ceph.mon.keyring
[search@node1 my-cluster]$ cat ceph.conf # 修改配置，添加public network
[global]
fsid = 57e12384-fd45-422b-bd0a-49da4149c1da
mon_initial_members = node1, node2, node3
mon_host = 10.0.4.14,10.0.4.6,10.0.4.15
auth_cluster_required = cephx
auth_service_required = cephx
auth_client_required = cephx
public network = 10.0.4.0/26 # add
- 安装ceph luminous(12.2.4)，参考 CREATE A CLUSTER
[search@node1 my-cluster]$ ceph-deploy install --release luminous node1 node2 node3
由于网速慢原因，安装可能会失败，我们可自行在各个节点手动安装 ceph
[search@node{1,2,3} ~]$ sudo yum install -y epel-release
[search@node{1,2,3} ~]$ sudo yum install -y ceph ceph-radosgw
[search@node1 ~]$ ceph --version
ceph version 12.2.4 (52085d5249a80c5f5121a76d6288429f35e4e77b) luminous (stable)
[search@node2 ~]$ ceph --version
ceph version 12.2.4 (52085d5249a80c5f5121a76d6288429f35e4e77b) luminous (stable)
[search@node3 ~]$ ceph --version
ceph version 12.2.4 (52085d5249a80c5f5121a76d6288429f35e4e77b) luminous (stable)
- 初始化 mon，参考 CREATE A CLUSTER
[search@node1 my-cluster]$ ceph-deploy mon create-initial
[search@node1 my-cluster]$ ls
ceph.bootstrap-mds.keyring ceph.bootstrap-osd.keyring ceph.client.admin.keyring ceph-deploy-ceph.log
ceph.bootstrap-mgr.keyring ceph.bootstrap-rgw.keyring ceph.conf ceph.mon.keyring
- 赋予各个节点使用命令免用户名权限
[search@node1 my-cluster]$ ceph-deploy admin node1 node2 node3
- 部署mgr(luminus+)
[search@node1 my-cluster]$ ceph-deploy mgr create node1
- 添加osd，参考 CREATE A CLUSTER
[search@node{1,2,3} ~]$ sudo fdisk -l # 每台 vm 均挂载 2块 20G SSD
Disk /dev/sda: 21.5 GB, 21474836480 bytes, 41943040 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk label type: dos
Disk identifier: 0x000b467e
Device Boot Start End Blocks Id System
/dev/sda1 * 2048 41943006 20970479+ 83 Linux
Disk /dev/sdb: 21.5 GB, 21474836480 bytes, 41943040 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
[search@node{1,2,3} ~]$ sudo lsblk -f
NAME FSTYPE LABEL UUID MOUNTPOINT
sda
└─sda1 xfs 6f15c206-f516-4ee8-a4b7-89ad880647db /
sdb
[search@node1 my-cluster]$ ceph-deploy osd create --data /dev/vdb node1
[search@node1 my-cluster]$ ceph-deploy osd create --data /dev/vdb node2
[search@node1 my-cluster]$ ceph-deploy osd create --data /dev/vdb node3
- 查看集群状态及osd tree
[search@node1 my-cluster]$ sudo ceph -s
cluster:
id: 57e12384-fd45-422b-bd0a-49da4149c1da
health: HEALTH_OK
services:
mon: 3 daemons, quorum node2,node1,node3
mgr: node1(active)
osd: 3 osds: 3 up, 3 in
data:
pools: 0 pools, 0 pgs
objects: 0 objects, 0 bytes
usage: 3164 MB used, 58263 MB / 61428 MB avail
pgs:
[search@node1 my-cluster]$ sudo ceph osd tree
ID CLASS WEIGHT TYPE NAME STATUS REWEIGHT PRI-AFF
-1 0.05846 root default
-3 0.01949 host node1
0 hdd 0.01949 osd.0 up 1.00000 1.00000
-5 0.01949 host node2
1 hdd 0.01949 osd.1 up 1.00000 1.00000
-7 0.01949 host node3
2 hdd 0.01949 osd.2 up 1.00000 1.00000
3.Dashboard配置
- 创建管理域密钥
[search@node1 my-cluster]$ sudo ceph auth get-or-create mgr.node1 mon 'allow profile mgr' osd 'allow *' mds 'allow *'
[mgr.node1]
key = AQC7os1ao1nFFhAANIB3V697rwKAHPc6ZiUPcw==
- 开启 ceph-mgr 管理域
[search@node1 my-cluster]$ sudo ceph-mgr -i node1
- 查看ceph状态，确认mgr状态为
active
[search@node1 my-cluster]$ sudo ceph -s
cluster:
id: 57e12384-fd45-422b-bd0a-49da4149c1da
health: HEALTH_OK
services:
mon: 3 daemons, quorum node2,node1,node3
mgr: node1(active, starting) # here
osd: 3 osds: 3 up, 3 in
data:
pools: 0 pools, 0 pgs
objects: 0 objects, 0 bytes
usage: 3164 MB used, 58263 MB / 61428 MB avail
pgs:
- 打开 dashboard 模块
[search@node1 my-cluster]$ sudo ceph mgr module enable dashboard
- 绑定开启 dashboard 模块的 ceph-mgr 节点的 ip 地址
[search@node1 my-cluster]$ sudo ceph config-key set mgr/dashboard/node1/server_addr 192.168.20.52
set mgr/dashboard/node1/server_addr
# dashboard 默认运行在7000端口
[search@node1 my-cluster]$ sudo netstat -tunpl | grep ceph-mgr # 确认7000端口服务运行
tcp 0 0 10.0.4.14:6800 0.0.0.0:* LISTEN 12499/ceph-mgr
tcp6 0 0 :::7000 :::* LISTEN 12499/ceph-mgr
- 打开
http://192.168.20.52:7000面板
dashboard.png
4.pool创建及使用
# 创建 pool
[search@node1 my-cluster]$ sudo ceph osd pool create rbd 128 128
[search@node1 my-cluster]$ sudo rbd pool init rbd
# 删除 pool
[search@node1 my-cluster]$ sudo ceph osd pool rm rbd rbd --yes-i-really-really-mean-it
Error EPERM: pool deletion is disabled; you must first set the mon_allow_pool_delete config option to true before you
can destroy a pool
[search@node1 my-cluster]$ cat ceph.conf
...
mon_allow_pool_delete = true
[search@node1 my-cluster]$ ceph-deploy --overwrite-conf config push node1
[search@node1 my-cluster]$ ceph-deploy --overwrite-conf config push node2
[search@node1 my-cluster]$ ceph-deploy --overwrite-conf config push node3
[search@node{1,2,3} ~]$ sudo systemctl restart ceph-mon.targe # 重启mon
[search@node1 my-cluster]$ sudo ceph osd pool rm rbd rbd --yes-i-really-really-mean-it
参考资料
如何在CentOS 7上修改主机名
Ceph Luminous（12.2.2）安装
centos7下搭建ceph luminous（12.2.1）--无网或网络较差
STORAGE CLUSTER QUICK START
DASHBOARD PLUGIN