基于cephfs搭建高可用分布式存储并mount到本地

 原文：https://www.fullstackmemo.com/2018/10/11/cephfs-ha-mount-storage/

服务器硬件配置及环境

项目说明
CPU 1核
内存 1GB
硬盘 40GB
系统 CentOS 7.5
时间同步服务 chrony
ceph 13.2.2-0

节点部署图

节点部署图

节点功能简介

项目说明
yum repo 如果部署环境每个节点都可以访问外网, 则无需做任何操作, 部署脚本会自动添加外网环境的yum源. 如果部署无法访问外网, 需要自行部署centos, epel, ceph三个yum源. 每个节点都必须可以访问所提到的所有yum源
时间同步服务器 每个节点都必须可以访问, 如果部署环境无法访问外网需要自行搭建时间同步服务器
client-x 需要挂载存储的设备, 需要同时可以访问每个storage-ha-x和yum源, 时间服务器
storage-deploy-1 用于统一部署ceph集群的工作机, 系统为CentOS 7.5
storage-ha-x 承载ceph各项服务的服务器节点, 系统为CentOS 7.5
mon Monitors, 节点映射管理, 身份验证管理, 需要达到冗余和高可用至少需要3个节点
osd object storage daemon, 对象存储服务, 需要达到冗余和高可用至少需要3个节点
mgr Manager, 用于跟踪运行指标和集群状态, 性能.
mds Metadata Serve, 提供cephfs的元数据存储

参考:

默认端口

项目说明
ssh tcp: 22
mon tcp: 6789
mds/mgr/osd tcp: 6800~7300

参考:

默认路径

项目说明
主配置文件 /etc/ceph/ceph.conf
配置文件夹 /etc/ceph
日志文件夹 /var/log/ceph
各服务认证key文件 /var/lib/ceph/{server name}/{hostname}/keyring
admin认证key文件 ceph.client.admin.keyring

部署脚本说明

  • node-init.sh: storage-ha-x节点初期运行的初始化脚本
  • admin-init.sh: storage-deploy-1节点初期运行的初始化脚本, 必须要在每个storage-ha-x节点都运行完node-init.sh之后才能运行.
  • ceph-deploy.sh: ceph部署脚本, 仅在storage-deploy-1节点上运行即可, 需要在node-init.sh和admin-init.sh运行完成且成功后运行.

PS: 脚本中涉及到的ip和其它不同信息请先自行修改后再运行.

脚本运行命令

请将 附录: 脚本内容章节或脚本Git库章节中的各个脚本放到各个对应服务器任意位置并使用以下命令按照顺序运行.

PS: 需严格按照部署脚本说明章节中的持续顺序执行脚本.

PS: 脚本中涉及到不同于当前环境的信息(如: ip, yum源, 密码, 主机名等)请先自行修改后再运行.

  • 初始化ceph节点
1
 
/bin/bash node-init.sh

执行命令结果

初始化ceph节点

  • 初始化部署节点
1
 
/bin/bash admin-init.sh

执行命令结果

初始化部署节点

初始化部署节点

  • 开始部署
1
 
/bin/bash ceph-deploy.sh

执行命令结果

部署结果-1

可以看到上方的pgs下方有个creating+peering, 这表示OSDs在创建和准备同步中.需要等待

这时可以在任意有admin角色的storage-ha-x节点上执行以下命令看查是否完成准备

1
 
ceph -s

当pgs显示为下图的active+clean代表各个节点同步完成.

部署结果-2

如果一直无法达到active+clean状态, 请参考以下操作文章:
TROUBLESHOOTING PGS

挂载存储

创建测试用户

以下命令在任意一个storage-ha-x服务器上运行

1
2
3
4
5
 
# 此命令含义是创建一个名为client.fs-test-1的用户, 对于挂载的根目录'/'只有可读权限, 对于挂载的'/test_1'目录有读写权限.
ceph fs authorize cephfs client.fs-test-1 / r /test_1 rw
# 命令输入完成后会返回类如下信息:
# [client.fs-test-1]
#         key = AQA0Cr9b9afRDBAACJ0M8HxsP41XmLhbSxWkqA==

获取用户授权信息

以下命令在任意一个添加过admin角色的storage-ha-x服务器上运行

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
 
# 获取client.admin用户的授权信息
ceph auth get client.admin
# 命令输入后会返回类似如下信息
# [client.admin]
#         key = AQAm4L5b60alLhAARxAgr9jQDLopr9fbXfm87w==
#         caps mds = "allow *"
#         caps mgr = "allow *"
#         caps mon = "allow *"
#         caps osd = "allow *"

# 获取client.fs-test-1用户的授权信息
ceph auth get client.fs-test-1
# 命令输入后会返回类似如下信息
# [client.fs-test-1]
#         key = AQA0Cr9b9afRDBAACJ0M8HxsP41XmLhbSxWkqA==
#         caps mds = "allow r, allow rw path=/test-1"
#         caps mon = "allow r"
#         caps osd = "allow rw tag cephfs data=cephfs"

挂载方式

挂载方式分为两种, 分别是cephfs和fuse. 选择其中一种方式进行挂载即可.

两种挂载方式的区别和优势请参考以下文章:
WHICH CLIENT?

cephfs方式

以下命令在任意需要挂载存储的client下执行

PS: 此挂载方式依赖于ceph, 需要先添加ceph和epel的yum源.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
 
# cephfs方式挂载
yum install ceph -y
mkdir -p /etc/ceph
mkdir -p /mnt/mycephfs
# 以下写入的secret请根据'获取用户授权信息'章节中获取到的'key'进行修改
cat > /etc/ceph/admin_secret.key << EOF
AQAm4L5b60alLhAARxAgr9jQDLopr9fbXfm87w==
EOF

# 以下写入的secret请根据'获取用户授权信息'章节中获取到的'key'进行修改
cat > /etc/ceph/test_cephfs_1_secret.key << EOF
AQA0Cr9b9afRDBAACJ0M8HxsP41XmLhbSxWkqA==
EOF

# 使用'admin'用户挂载cephfs的根目录
# ip或主机名请根据实际情况修改
# 这里填写的'name=admin'是'client.admin'点后面的'admin'.
mount.ceph 192.168.60.111:6789,192.168.60.112:6789,192.168.60.113:6789:/ /mnt/mycephfs -o name=admin,secretfile=/etc/ceph/admin_secret.key

# 使用只读的用户挂载
mkdir -p /mnt/mycephfs/test_1
mkdir -p /mnt/test_cephfs_1
# 使用'fs-test-1'用户挂载cephfs的根目录
# ip或主机名请根据实际情况修改
# 这里填写的'name=fs-test-1'是'client.fs-test-1'点后面的'fs-test-1'.
mount.ceph 192.168.60.111:6789,192.168.60.112:6789,192.168.60.113:6789:/ /mnt/test_cephfs_1 -o name=fs-test-1,secretfile=/etc/ceph/test_cephfs_1_secret.key

# 开机自动挂载
cat >> /etc/fstab << EOF
192.168.60.111:6789,192.168.60.112:6789,192.168.60.113:6789:/     /mnt/mycephfs    ceph    name=admin,secretfile=/etc/ceph/secret.key,noatime,_netdev    0       2
EOF

fuse方式

以下命令在任意需要挂载存储的client下执行

PS: 此挂载方式依赖于ceph-fuse, 需要先添加ceph和epel的yum源.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
 
yum install ceph-fuse -y
mkdir -p /etc/ceph
mkdir -p /mnt/mycephfs

# 获取storage-ha-x任意一个节点上的ceph配置文件
scp storage@storage-ha-1:/etc/ceph/ceph.conf /etc/ceph/ceph.conf

# 以下写入的secret请根据'获取用户授权信息'章节中获取到的'key'进行修改
cat > /etc/ceph/ceph.keyring << EOF
[client.admin]
        key = AQAm4L5b60alLhAARxAgr9jQDLopr9fbXfm87w==
        caps mds = "allow *"
        caps mgr = "allow *"
        caps mon = "allow *"
        caps osd = "allow *"
[client.fs-test-1]
        key = AQA0Cr9b9afRDBAACJ0M8HxsP41XmLhbSxWkqA==
        caps mds = "allow r, allow rw path=/test-1"
        caps mon = "allow r"
        caps osd = "allow rw tag cephfs data=cephfs"
EOF

# 使用'admin'用户挂载cephfs的根目录
# ip或主机名请根据实际情况修改
ceph-fuse -m 192.168.60.111:6789,192.168.60.112:6789,192.168.60.113:6789 /mnt/mycephfs
# 开机自动挂载
cat >> /etc/fstab << EOF
none    /mnt/ceph  fuse.ceph ceph.id=admin,ceph.conf=/etc/ceph/ceph.conf,_netdev,defaults  0 0
EOF

# 使用只读的用户挂载
mkdir -p /mnt/mycephfs/test_1
mkdir -p /mnt/test_cephfs_1
# 使用'fs-test-1'用户挂载cephfs的根目录
# ip或主机名请根据实际情况修改
# 这里填写的'-n client.fs-test-1'是完整的'client.fs-test-1'.
ceph-fuse -m 192.168.60.111:6789,192.168.60.112:6789,192.168.60.113:6789 -n client.fs-test-1 /mnt/test_cephfs_1

挂载结果

挂载结果可以使用以下命令查看

1
 
df -h

挂载结果

运维命令

  • 状态查看
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
 
# 查看集群整体状态
ceph -s

# 查看集群健康状态
ceph health

# 查看集群健康状态详情
ceph health detail

# 查看cephfs列表
ceph fs ls

# 查看mds状态
ceph mds stat

# 查看 osd节点状态
ceph osd tree

# 查看监视器情况
ceph quorum_status --format json-pretty
  • 简单写性能测试
1
2
 
# 在挂载了存储的client下简单测试写性能
time dd if=/dev/zero of=/mnt/mycephfs/test.dbf bs=8k count=3000 oflag=direct

测试结果

  • 删除cephfs和pool
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
 
# 删除cephfs前需要的操作
# 停止每个mds节点的mds服务
# 每个mds节点上都要执行
systemctl stop ceph-mds.target

# 仅在任意一台'storage-ha-x'上执行
# 删除cephfs
ceph fs rm cephfs --yes-i-really-mean-it

# 删除pool
# 需要删除pool的时候需要写两次pool名外带'--yes-i-really-really-mean-it'参数
ceph osd pool rm cephfs_data cephfs_data --yes-i-really-really-mean-it
ceph osd pool rm cephfs_metadata cephfs_metadata --yes-i-really-really-mean-it

# 停止每个mds节点的mds服务
# 每个mds节点上都要执行
systemctl start ceph-mds.target
  • 同步ceph配置文件
1
2
3
4
5
6
7
8
9
10
11
 
# 同步的配置文件
# 如果节点上有配置文件且和当前要同步的配置文件不通, 需要带'--overwrite-conf'参数
# 此命令会把执行此命令目录下的'ceph.conf'文件同步到各个指定节点上
ceph-deploy --overwrite-conf config push storage-ha-1 storage-ha-2 storage-ha-3

# 重启每个节点的cepf相关服务
# 需要在有对应功能节点的节点上分别运行以下命令
systemctl restart ceph-osd.target
systemctl restart ceph-mds.target
systemctl restart ceph-mon.target
systemctl restart ceph-mgr.target

FAQ

  • Q: health_warn：clock skew detected on mon
    A: 使用chrony同步每台服务器节点的时间

  • Q: Error ERANGE: pg_num “*“ size “*“ would mean “*“ total pgs, which exceeds max “*“ (mon_max_pg_per_osd 250 num_in_osds “\“)
    A: ceph.conf配置文件中加入mon_max_pg_per_osd = 1000(参数中的数值自己根据实际情况修改)并用同步ceph配置文件方式上传到各个节点, 并重启ceph-mon.target

  • Q: too many PGs per OSD
    A: ceph.conf配置文件中加入mon_max_pg_per_osd = 1000(参数中的数值自己根据实际情况修改)并用同步ceph配置文件方式上传到各个节点, 并重启ceph-mon.target

参考

ceph cephx认证参考
设置cephfs访问权限
ceph用户管理
ceph-fuse方式挂载
Ceph 运维手册
Red Hat Ceph存储—《深入理解Ceph架构》
Ceph常规操作及常见问题梳理

脚本Git库

https://github.com/x22x22/cephfs-verify-script

附录: 脚本内容

  • node-init.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
 
#!/bin/bash

# 禁用ipv6, 加大pid限制
cat >>/etc/sysctl.conf <<EOF
net.ipv6.conf.all.disable_ipv6 = 1
net.ipv6.conf.default.disable_ipv6 = 1
kernel.pid_max = 4194303
EOF

sysctl -p
sysctl -w net.ipv6.conf.all.disable_ipv6=1
sysctl -w net.ipv6.conf.default.disable_ipv6=1

# 简单代替dns服务器写入当前环境中的主机名和ip的对应关系
cat >>/etc/hosts <<EOF

192.168.60.110 storage-deploy-1
192.168.60.111 storage-ha-1
192.168.60.112 storage-ha-2
192.168.60.113 storage-ha-3
EOF

systemctl stop firewalld
systemctl disable firewalld
setenforce 0
sed -i 's/SELINUX=.*/SELINUX=disabled/g' /etc/selinux/config

# 添加一个storage用户, 用于ceph-deploy工具进行节点的安装和操作
useradd -d /home/storage -m storage
echo 'fullstackmemo***' | passwd --stdin storage
echo "storage ALL = (root) NOPASSWD:ALL" | tee /etc/sudoers.d/storage
chmod 0440 /etc/sudoers.d/storage

# 添加ceph的yum源, 如果无法访问外网请自行搭建并修改
cat >/etc/yum.repos.d/ceph.repo <<'EOF'
[Ceph]
name=Ceph packages for $basearch
baseurl=http://mirror.tuna.tsinghua.edu.cn/ceph/rpm-mimic/el7/$basearch
enabled=1
gpgcheck=1
type=rpm-md
gpgkey=https://mirror.tuna.tsinghua.edu.cn/ceph/keys/release.asc
priority=1

[Ceph-noarch]
name=Ceph noarch packages
baseurl=http://mirror.tuna.tsinghua.edu.cn/ceph/rpm-mimic/el7/noarch
enabled=1
gpgcheck=1
type=rpm-md
gpgkey=https://mirror.tuna.tsinghua.edu.cn/ceph/keys/release.asc
priority=1

[ceph-source]
name=Ceph source packages
baseurl=http://mirror.tuna.tsinghua.edu.cn/ceph/rpm-mimic/el7/SRPMS
enabled=1
gpgcheck=1
type=rpm-md
gpgkey=https://mirror.tuna.tsinghua.edu.cn/ceph/keys/release.asc
priority=1

EOF

mv /etc/yum.repos.d/CentOS-Base.repo /etc/yum.repos.d/CentOS-Base.repo.backup

# 修改CentOS的yum基础源, 如果无法访问外网请自行搭建并修改
cat >/etc/yum.repos.d/CentOS-Base.repo <<'EOF'
# CentOS-Base.repo
#
# The mirror system uses the connecting IP address of the client and the
# update status of each mirror to pick mirrors that are updated to and
# geographically close to the client.  You should use this for CentOS updates
# unless you are manually picking other mirrors.
#
# If the mirrorlist= does not work for you, as a fall back you can try the
# remarked out baseurl= line instead.
#
#

[base]
name=CentOS-$releasever - Base
baseurl=https://mirrors.tuna.tsinghua.edu.cn/centos/$releasever/os/$basearch/
#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=os
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7

#released updates
[updates]
name=CentOS-$releasever - Updates
baseurl=https://mirrors.tuna.tsinghua.edu.cn/centos/$releasever/updates/$basearch/
#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=updates
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7

#additional packages that may be useful
[extras]
name=CentOS-$releasever - Extras
baseurl=https://mirrors.tuna.tsinghua.edu.cn/centos/$releasever/extras/$basearch/
#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=extras
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7

#additional packages that extend functionality of existing packages
[centosplus]
name=CentOS-$releasever - Plus
baseurl=https://mirrors.tuna.tsinghua.edu.cn/centos/$releasever/centosplus/$basearch/
#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=centosplus
gpgcheck=1
enabled=0
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
EOF

yum makecache fast
# 安装CentOS的yum epel源
yum install -y epel-release

# 修改CentOS的yum epel源, 如果无法访问外网请自行搭建并修改
cat >/etc/yum.repos.d/epel.repo <<'EOF'
[epel]
name=Extra Packages for Enterprise Linux 7 - $basearch
baseurl=https://mirrors.tuna.tsinghua.edu.cn/epel/7/$basearch
#mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-7&arch=$basearch
failovermethod=priority
enabled=1
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7

[epel-debuginfo]
name=Extra Packages for Enterprise Linux 7 - $basearch - Debug
baseurl=https://mirrors.tuna.tsinghua.edu.cn/epel/7/$basearch/debug
#mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-debug-7&arch=$basearch
failovermethod=priority
enabled=0
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7
gpgcheck=1

[epel-source]
name=Extra Packages for Enterprise Linux 7 - $basearch - Source
baseurl=https://mirrors.tuna.tsinghua.edu.cn/epel/7/SRPMS
#mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-source-7&arch=$basearch
failovermethod=priority
enabled=0
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7
gpgcheck=1
EOF

yum makecache
yum install yum-plugin-priorities chrony parted xfsprogs -y
mv /etc/chrony.conf /etc/chrony.conf.bk

# 添加时间同步服务器, 如果无法访问外网请自行搭建并修改
# 添加时间同步服务器, 如果无法访问外网请更换成yum.yfb.sunline.cn和nexus.yfb.sunline.cn
cat > /etc/chrony.conf << EOF
server ntp.api.bz iburst
driftfile /var/lib/chrony/drift
makestep 1.0 3
rtcsync
logdir /var/log/chrony
EOF

systemctl enable chronyd
systemctl restart chronyd
sleep 10
chronyc activity
chronyc sources -v
hwclock -w

# 这里将/dev/sdb作为ceph的存储池, 所以先格式化/dev/sdb, 请根据自己实际情况修改
parted -s /dev/sdb mklabel gpt mkpart primary xfs 0% 100%
partprobe /dev/sdb
mkfs.xfs /dev/sdb -f
  • admin-init.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
 
#!/bin/bash

# 禁用ipv6
cat >>/etc/sysctl.conf <<EOF
net.ipv6.conf.all.disable_ipv6 = 1
net.ipv6.conf.default.disable_ipv6 = 1
EOF

sysctl -p
sysctl -w net.ipv6.conf.all.disable_ipv6=1
sysctl -w net.ipv6.conf.default.disable_ipv6=1

# 简单代替dns服务器写入当前环境中的主机名和ip的对应关系
cat >>/etc/hosts <<EOF

192.168.60.110 storage-deploy-1
192.168.60.111 storage-ha-1
192.168.60.112 storage-ha-2
192.168.60.113 storage-ha-3
EOF

systemctl stop firewalld
systemctl disable firewalld
setenforce 0
sed -i 's/SELINUX=.*/SELINUX=disabled/g' /etc/selinux/config

# 添加ceph的yum源, 如果无法访问外网请自行搭建并修改
cat >/etc/yum.repos.d/ceph.repo <<'EOF'
[Ceph]
name=Ceph packages for $basearch
baseurl=http://mirror.tuna.tsinghua.edu.cn/ceph/rpm-mimic/el7/$basearch
enabled=1
gpgcheck=1
type=rpm-md
gpgkey=https://mirror.tuna.tsinghua.edu.cn/ceph/keys/release.asc
priority=1

[Ceph-noarch]
name=Ceph noarch packages
baseurl=http://mirror.tuna.tsinghua.edu.cn/ceph/rpm-mimic/el7/noarch
enabled=1
gpgcheck=1
type=rpm-md
gpgkey=https://mirror.tuna.tsinghua.edu.cn/ceph/keys/release.asc
priority=1

[ceph-source]
name=Ceph source packages
baseurl=http://mirror.tuna.tsinghua.edu.cn/ceph/rpm-mimic/el7/SRPMS
enabled=1
gpgcheck=1
type=rpm-md
gpgkey=https://mirror.tuna.tsinghua.edu.cn/ceph/keys/release.asc
priority=1

EOF

mv /etc/yum.repos.d/CentOS-Base.repo /etc/yum.repos.d/CentOS-Base.repo.backup

# 修改CentOS的yum基础源, 如果无法访问外网请自行搭建并修改
cat >/etc/yum.repos.d/CentOS-Base.repo <<'EOF'
# CentOS-Base.repo
#
# The mirror system uses the connecting IP address of the client and the
# update status of each mirror to pick mirrors that are updated to and
# geographically close to the client.  You should use this for CentOS updates
# unless you are manually picking other mirrors.
#
# If the mirrorlist= does not work for you, as a fall back you can try the
# remarked out baseurl= line instead.
#
#

[base]
name=CentOS-$releasever - Base
baseurl=https://mirrors.tuna.tsinghua.edu.cn/centos/$releasever/os/$basearch/
#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=os
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7

#released updates
[updates]
name=CentOS-$releasever - Updates
baseurl=https://mirrors.tuna.tsinghua.edu.cn/centos/$releasever/updates/$basearch/
#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=updates
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7

#additional packages that may be useful
[extras]
name=CentOS-$releasever - Extras
baseurl=https://mirrors.tuna.tsinghua.edu.cn/centos/$releasever/extras/$basearch/
#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=extras
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7

#additional packages that extend functionality of existing packages
[centosplus]
name=CentOS-$releasever - Plus
baseurl=https://mirrors.tuna.tsinghua.edu.cn/centos/$releasever/centosplus/$basearch/
#mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=centosplus
gpgcheck=1
enabled=0
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
EOF

yum makecache fast
# 安装CentOS的yum epel源
yum install -y epel-release

# 修改CentOS的yum epel源, 如果无法访问外网请自行搭建并修改
cat >/etc/yum.repos.d/epel.repo <<'EOF'
[epel]
name=Extra Packages for Enterprise Linux 7 - $basearch
baseurl=https://mirrors.tuna.tsinghua.edu.cn/epel/7/$basearch
#mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-7&arch=$basearch
failovermethod=priority
enabled=1
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7

[epel-debuginfo]
name=Extra Packages for Enterprise Linux 7 - $basearch - Debug
baseurl=https://mirrors.tuna.tsinghua.edu.cn/epel/7/$basearch/debug
#mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-debug-7&arch=$basearch
failovermethod=priority
enabled=0
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7
gpgcheck=1

[epel-source]
name=Extra Packages for Enterprise Linux 7 - $basearch - Source
baseurl=https://mirrors.tuna.tsinghua.edu.cn/epel/7/SRPMS
#mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-source-7&arch=$basearch
failovermethod=priority
enabled=0
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7
gpgcheck=1
EOF

yum makecache
yum install yum-plugin-priorities chrony sshpass ceph-deploy ceph -y
mv /etc/chrony.conf /etc/chrony.conf.bk

# 添加时间同步服务器, 如果无法访问外网请自行搭建并修改
cat > /etc/chrony.conf << EOF
server ntp.api.bz iburst
driftfile /var/lib/chrony/drift
makestep 1.0 3
rtcsync
logdir /var/log/chrony
EOF

systemctl enable chronyd
systemctl restart chronyd
sleep 10
chronyc activity
chronyc sources -v
hwclock -w

rm -f "${HOME}"/.ssh/ceph_id_rsa
ssh-keygen -t rsa -b 4096 -f "${HOME}"/.ssh/ceph_id_rsa -N ''
cat >"${HOME}"/.ssh/config <<EOF
Host storage-ha-1
   Hostname storage-ha-1
   User storage
   IdentityFile ${HOME}/.ssh/ceph_id_rsa
   IdentitiesOnly yes
   StrictHostKeyChecking no
Host storage-ha-2
   Hostname storage-ha-2
   User storage
   IdentityFile ${HOME}/.ssh/ceph_id_rsa
   IdentitiesOnly yes
   StrictHostKeyChecking no
Host storage-ha-3
   Hostname storage-ha-3
   User storage
   IdentityFile ${HOME}/.ssh/ceph_id_rsa
   IdentitiesOnly yes
   StrictHostKeyChecking no
EOF
chmod 0400 "${HOME}"/.ssh/config
sshpass -p "fullstackmemo***" ssh-copy-id -i ~/.ssh/ceph_id_rsa storage@storage-ha-1
sshpass -p "fullstackmemo***" ssh-copy-id -i ~/.ssh/ceph_id_rsa storage@storage-ha-2
sshpass -p "fullstackmemo***" ssh-copy-id -i ~/.ssh/ceph_id_rsa storage@storage-ha-3

mkdir -p "${HOME}"/ceph-cluster
cd "${HOME}"/ceph-cluster || exit
  • ceph-deploy.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
 
#!/bin/bash

mkdir -p "${HOME}"/ceph-cluster
cd "${HOME}"/ceph-cluster || exit
ceph-deploy new storage-ha-1 storage-ha-2 storage-ha-3

cat >>ceph.conf <<EOF
# 'public network':
# 整个集群所存在的网段
# 这里需要根据实际情况修改
public network = 192.168.60.0/24
osd pool default size = 3
osd pool default min size = 2
osd pool default pg num = 100
osd pool default pgp num = 100
# 'mon allow pool delete': 
# 此设置允许删除pool的操作, poc环境为方便操作加上此选项, 生产环境建议注释
mon allow pool delete = true

[osd]
osd_max_backfills = 1
osd_recovery_max_active = 1
osd_recovery_op_priority = 1
EOF

# 在各个节点上安装ceph, 并指定了外网的ceph yum源, 如果无法访问外网请自行搭建并修改
ceph-deploy install storage-ha-1 storage-ha-2 storage-ha-3 --repo-url http://mirrors.ustc.edu.cn/ceph/rpm-mimic/el7 --gpg-url 'http://mirrors.ustc.edu.cn/ceph/keys/release.asc'
# 初始化mon服务和key信息
ceph-deploy mon create-initial
ceph-deploy mon add storage-ha-2
ceph-deploy mon add storage-ha-3
ceph-deploy admin storage-ha-1 storage-ha-2 storage-ha-3
ceph-deploy mgr create storage-ha-1 storage-ha-2 storage-ha-3

# 添加存储服务节点上的裸盘到存储池中
ceph-deploy osd create --data /dev/sdb storage-ha-1
ceph-deploy osd create --data /dev/sdb storage-ha-2
ceph-deploy osd create --data /dev/sdb storage-ha-3

ceph-deploy mds create storage-ha-1 storage-ha-2 storage-ha-3

ssh storage@storage-ha-1 << EOF
# 创建两个pool, 服务于cephfs, cephfs至少需要两个pool, 分别做metadata和data
sudo ceph osd pool create cephfs_data 100
# 使用raid 5方式存储数据即erasure类型, 当单个文件平均大小大于8k时erasure比replicated有优势.
# sudo ceph osd pool create cephfs_data 100 100 erasure
# sudo ceph osd pool set cephfs_data allow_ec_overwrites true
# sudo metadata pool必须使用replicated类型.
sudo ceph osd pool create cephfs_metadata 100
# 如果使用了erasure类型, 此步骤跳过
sudo ceph osd pool set cephfs_data size 3

sudo ceph osd pool set cephfs_metadata size 3
sudo ceph fs new cephfs cephfs_metadata cephfs_data

# 查看集群各项信息
sudo ceph quorum_status --format json-pretty
sudo ceph fs ls
sudo ceph mds stat
sudo ceph health
sudo ceph -s
EOF
posted @ 2019-10-31 18:56  sunsky303  阅读(900)  评论(0)  编辑  收藏  举报
显示搜索