ceph安装

使用ceph-deploy部署部署版本ceph-luminous

 

一，下载安装包，制作yum源，yum将挂载到192.168.100.100

（1）下载安装包

mkdir ceph-luminous

repotrack ceph ceph-mgr ceph-mon ceph-mds ceph-osd ceph-fuse ceph-radosgw -p /root/ceph-luminous/

repotrack ceph-deploy -p /root/ceph-luminous/

repotrack fcgi -p /root/ceph-luminous

repotrack fcgi-devel -p /root/ceph-luminous

repotrack epel-release -p /root/ceph-luminous

repotrack yum-plugin-priorities -p /root/ceph-luminous

repotrack python2-pip -p /root/ceph-luminous

（2）制作yum源

安装httpd服务 yum install httpd

将安装包移动到/var/www/html/ceph目录下，启动http服务后就可以访问

下载https://mirrors.aliyun.com/ceph/keys/release.asc到/var/www/html/文件夹下，ceph-deploy安装时会找key文件

/var/www/html/ceph

├── noarch

│   └── repodata

├── SRPMS

│   └── repodata

└── x86_64

├── repodata

注意：

ceph安装包都在x86_64目录下

安装过程会找noarch和SRPMS目录，因此需要创建noarch和SRPMS目录（空目录），使用createrepo命令在noarch和SRPMS目录下生成 repodata目录，不然安装会报错

 

 

二，ceph节点部署

 

2.1 主机信息

 

主机名	操作系统	配置	CEPH组件	私网IP	用户名/密码
yum	CentOS7.6 64bit	2C2G sda:50G		192.168.100.100
ceph1	CentOS7.6 64bit	2C2G sda:50G sdb:2G	admin mon osd	192.168.100.131	root/pwd@123
ceph2	CentOS7.6 64bit	2C2G sda:50G sdb:2G	mon osd	192.168.100.132	root/pwd@123
ceph3	CentOS7.6 64bit	2C2G sda:50G sdb:2G	mon osd	192.168.100.133	root/pwd@123

 

2.2 配置各主机hosts文件

 

# cat /etc/hosts

127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4

::1         localhost localhost.localdomain localhost6 localhost6.localdomain6

192.168.100.131 ceph1

192.168.100.132 ceph2

192.168.100.133 ceph3

 

2.3 ceph1免密登录ceph2/ceph3

 

[root@ceph1 ~]# ssh-keygen

[root@ceph1 ~]# ssh-copy-id -i .ssh/id_rsa.pub ceph2

[root@ceph1 ~]# ssh-copy-id -i .ssh/id_rsa.pub ceph3

 

2.4 配置各节点yum源

 

# cat /etc/yum.repos.d/base.repo

[ceph]

name=ceph

baseurl=http://192.168.100.100/ceph/x86_64

gpgcheck=0

 

[base]

name=base

baseurl=http://192.168.100.100/centos/7/os/x86_64

gpgcheck=0

 

2.5 ceph1节点安装ceph-deploy工具

 

yum install ceph-deploy

 

2.6 创建集群

创建集群，并设置ceph1 ceph2 ceph3为 mon节点

[root@ceph1 ~]# mkdir /etc/ceph && cd /etc/ceph 生成配置文件在当前目录下

[root@ceph1 ~]# ceph-deploy new ceph1 ceph2 ceph3

[root@ceph1 ~]# cat /etc/ceph/ceph.conf

[global]

fsid = ba902222-4795-4e07-90a3-b1953928d187

mon_initial_members = ceph1, ceph2, ceph3

mon_host = 192.168.100.131,192.168.100.132,192.168.100.133

auth_cluster_required = cephx

auth_service_required = cephx

auth_client_required = cephx

 

public network = 192.168.100.0/24

 

2.7 所有节点安装ceph

 

# ceph-deploy install ceph1 ceph2 ceph3

[root@ceph1 ~]# ceph --version

ceph version 12.2.12 (1436006594665279fe734b4c15d7e08c13ebd777) luminous (stable)

 

2.8 激活monitor服务

 

[root@ceph1 ceph]#ceph-deploy mon create-initial

 

 

2.9 统一配置

 

用ceph-deploy把配置文件和admin密钥拷贝到所有节点，这样每次执行Ceph命令行时就无需指定monitor地址和ceph.client.admin.keyring了

[root@ceph1 ceph]# ceph-deploy admin ceph1 ceph2 ceph3

ceph-deploy --overwrite-conf admin node{1..4} 同步配置文件

 

2.10 部署OSD服务

 

yum install python2-pip

[root@ceph1 ceph]# ceph-volume lvm create --data /dev/sdb

[root@ceph2 ceph]# ceph-volume lvm create --data /dev/sdb

[root@ceph3 ceph]# ceph-volume lvm create --data /dev/sdb