本文记录了在 CentOS 7.2 环境下安装 Ceph v9.2.0 的步骤。
安装环境
- 3 台 CentOS 7.2 虚机，内核号：3.10.0-327.el7.x86_64
- 各分配两块硬盘，/dev/sdb 作 osd，共计 3 个 osd，3 个 mon
服务划分如下：
| hostname | ip | servers |
|---|
| ceph-1 | 10.200.175.11 | mon.0 & osd.0 |
| ceph-2 | 10.200.175.12 | mon.1 & osd.1 |
| ceph-3 | 10.200.175.13 | mon.2 & osd.2 |
安装步骤
1. 下载 rpm 包
http://download.ceph.com/rpm-infernalis/el7/x86_64/
2. 制作本地 rpm 源
1
2
3
4
5
6
7
8
9
| # yum install createrepo -y
# createrepo -v /root/cephrpm/
# cat >> /etc/yum.repos.d/local.repo << EOF
[local_server]
name = Local Repo
baseurl = file:///root/cephrpm/
enabled = 1
gpgcheck = 0
EOF
|
3. 安装
1
2
| # yum clean all
# yum install ceph ceph-common -y
|
4. leveldb
1
| # wget ftp://rpmfind.net/linux/fedora/linux/releases/22/Everything/x86_64/os/Packages/l/leveldb-1.12.0-6.fc21.x86_64.rpm
|
5. 配置 ceph.conf
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
| # cat >> /etc/ceph/ceph.conf << EOF
[global]
auth cluster required = none
auth service required = none
auth client required = none
keyring = /etc/ceph/keyring
cluster network = 10.200.175.1/23
public network = 10.200.175.1/23
osd pool default pg num = 1024
osd pool default pgp num = 1024
osd pool default size = 2
fsid = a3972be0-382d-4227-a3d0-c6e98c18b818
[mon]
mon data = /var/lib/ceph/mon/mon.$id
[osd]
osd journal size = 4096
filestore flusher = false
osd data = /var/lib/ceph/osd/ceph-$id
osd journal = /var/lib/ceph/osd/ceph-$id/journal
osd mkfs type = xfs
osd mkfs options xfs = -f
osd mount options xfs = rw,noatime
keyring = /var/lib/ceph/osd/ceph-$id/keyring
[mon.0]
host = ceph-1
mon addr = 10.200.175.11:6789
[mon.1]
host = ceph-2
mon addr = 10.200.175.12:6789
[mon.2]
host = ceph-3
mon addr = 10.200.175.13:6789
[osd.0]
host = ceph-1
devs = /dev/sdb1
cluster addr = 10.200.175.11
public addr = 10.200.175.11
[osd.1]
host = ceph-2
devs = /dev/sdb1
cluster addr = 10.200.175.12
public addr = 10.200.175.12
[osd.2]
host = ceph-3
devs = /dev/sdb1
cluster addr = 10.200.175.13
public addr = 10.200.175.13
EOF
|
6. 启动 mon
1
2
| # ceph-mon --mkfs –i 0
# ceph-mon -i 0 -c /etc/ceph/ceph.conf
|
7. 配置 osd
1
2
3
4
5
6
7
8
| # parted -a optimal --script /dev/sdb mktable gpt
# parted -a optimal -s /dev/sdb mkpart ceph 0% 100%
# mkfs.xfs /dev/sdb1
# blkid # get block device id of /dev/sdb1
# ceph osd create $BLOCKID
# mkdir /var/lib/ceph/osd/ceph-$ID
# mount -o rw,noatime,inode64 /dev/sdb1 /var/lib/ceph/osd/ceph-$ID
# ceph-osd -c /etc/ceph/ceph.conf -i $ID --mkfs --osd-uuid $BLOCKID
|
8. 启动 osd
1
| # ceph-osd -c /etc/ceph/ceph.conf -i $ID
|
9. crushmap
1
2
3
4
5
6
| # ceph osd getcrushmap -o {compiled-crushmap-filename}
# crushtool -d {compiled-crushmap-filename} -o {decompiled-crushmap-filename}
# edit {decompiled-crushmap-filename}
# ...
# crushtool -c {decompiled-crush-map-filename} -o {compiled-crush-map-filename}
# ceph osd setcrushmap -i {compiled-crushmap-filename}
|
10. 查看状态
1
2
3
| # ceph -s
# ceph osd tree
# rados lspools
|