基于docker的ceph集群

ceph部署环境

主机名称	内网IP
node01	192.168.0.1
node02	192.168.0.2
node03	192.168.0.3

虚机开启selinux：

vi /etc/selinux/config

设置：

SELINUX=enforcing

设置目录权限：

mkdir -p /etc/ceph
mkdir -p /var/lib/ceph
chcon -Rt svirt_sandbox_file_t /etc/ceph
chcon -Rt svirt_sandbox_file_t /var/lib/ceph

如果系统只有一个数据盘，并且数据盘已经分区，需要对数据盘进行分区删除，并进行格式化处理。
格式化命令如下：

mkfs.ext4 /dev/vdb

下面我们采用docker进行安装，ceph/daemon的版本为tag-build-master-jewel-centos-7。

node01启动monitor

启动monitor：

docker run -d --name=mon --net=host \
--restart=always \
-v /etc/ceph:/etc/ceph \
-v /var/lib/ceph:/var/lib/ceph \
-e MON_IP=192.168.0.1 \
-e CEPH_PUBLIC_NETWORK=192.168.0.0/24 \
ceph/daemon:tag-build-master-jewel-centos-7 mon

查看ceph状态：

docker exec b3cc55582498 ceph -s

执行结果：

# docker exec b3cc55582498 ceph -s
    cluster 05527e2d-5d80-4c85-8d35-7dcddafa197e
     health HEALTH_ERR
            no osds
     monmap e1: 1 mons at {iZbp1isotv99f45cg37lxgZ=192.168.0.1:6789/0}
            election epoch 3, quorum 0 iZbp1isotv99f45cg37lxgZ
     osdmap e1: 0 osds: 0 up, 0 in
            flags sortbitwise,require_jewel_osds
      pgmap v2: 64 pgs, 1 pools, 0 bytes data, 0 objects
            0 kB used, 0 kB / 0 kB avail
                  64 creating

拷贝node01配置文件到其他机器

scp -r /etc/ceph/ceph* root@192.168.0.2:/etc/ceph/
scp -r /var/lib/ceph/bootstrap-* root@192.168.0.2:/var/lib/ceph/

scp -r /etc/ceph/ceph* root@192.168.0.3:/etc/ceph/
scp -r /var/lib/ceph/bootstrap-* root@192.168.0.3:/var/lib/ceph/

其他节点启动monitor

在node02上执行：

docker run -d --name=mon --net=host \
--restart=always \
-v /etc/ceph:/etc/ceph \
-v /var/lib/ceph:/var/lib/ceph \
-e MON_IP=192.168.0.2 \
-e CEPH_PUBLIC_NETWORK=192.168.0.0/24 \
ceph/daemon:tag-build-master-jewel-centos-7 mon

在node03上执行：

docker run -d --name=mon --net=host \
--restart=always \
-v /etc/ceph:/etc/ceph \
-v /var/lib/ceph:/var/lib/ceph \
-e MON_IP=192.168.0.3 \
-e CEPH_PUBLIC_NETWORK=192.168.0.0/24 \
ceph/daemon:tag-build-master-jewel-centos-7 mon

查看ceph状态：

docker exec b3cc55582498 ceph -s

执行结果：

#docker exec b3cc55582498 ceph -s
    cluster 230a93b6-3876-4d69-bffe-c0c054c49653
     health HEALTH_ERR
            no osds
     monmap e3: 3 mons at {iZbp1isotv99f45cg37lxfZ=192.168.0.1:6789/0,iZbp1isotv99f45cg37lxiZ=192.168.0.2:6789/0,iZbp1isotv99f45cg37lxjZ=192.168.0.3:6789/0}
            election epoch 6, quorum 0,1,2 iZbp1isotv99f45cg37lxjZ,iZbp1isotv99f45cg37lxfZ,iZbp1isotv99f45cg37lxiZ
     osdmap e1: 0 osds: 0 up, 0 in
            flags sortbitwise,require_jewel_osds
      pgmap v2: 64 pgs, 1 pools, 0 bytes data, 0 objects
            0 kB used, 0 kB / 0 kB avail
                  64 creating

在所有节点部署osd

docker run -d --name=osd1 --net=host \
--restart=always \
-v /etc/ceph:/etc/ceph \
-v /var/lib/ceph/:/var/lib/ceph/ \
-v /dev/:/dev/ --privileged=true \
-e OSD_FORCE_ZAP=1 \
-e OSD_DEVICE=/dev/vdb \
ceph/daemon:tag-build-master-jewel-centos-7 osd_ceph_disk

查看ceph状态：

docker exec b3cc55582498 ceph -s

执行结果：

# docker exec b3cc55582498 ceph -s
    cluster 230a93b6-3876-4d69-bffe-c0c054c49653
     health HEALTH_ERR
            37 pgs are stuck inactive for more than 300 seconds
            27 pgs degraded
            3 pgs peering
            37 pgs stuck inactive
            64 pgs stuck unclean
            27 pgs undersized
     monmap e3: 3 mons at {iZbp1isotv99f45cg37lxfZ=192.168.0.1:6789/0,iZbp1isotv99f45cg37lxiZ=192.168.0.2:6789/0,iZbp1isotv99f45cg37lxjZ=192.168.0.3:6789/0}
            election epoch 6, quorum 0,1,2 iZbp1isotv99f45cg37lxjZ,iZbp1isotv99f45cg37lxfZ,iZbp1isotv99f45cg37lxiZ
     osdmap e10: 3 osds: 3 up, 3 in
            flags sortbitwise,require_jewel_osds
      pgmap v11: 64 pgs, 1 pools, 0 bytes data, 0 objects
            35320 kB used, 699 GB / 699 GB avail
                  34 creating
                  27 active+undersized+degraded
                   3 creating+peering

在node01部署mds

docker run -d --name=mds --net=host \
--restart=always \
-v /etc/ceph:/etc/ceph \
-v /var/lib/ceph/:/var/lib/ceph/ \
-e CEPHFS_CREATE=1 \
ceph/daemon:tag-build-master-jewel-centos-7 mds

查看ceph状态：

docker exec b3cc55582498 ceph -s

执行结果：

# docker exec b3cc55582498 ceph -s
    cluster 230a93b6-3876-4d69-bffe-c0c054c49653
     health HEALTH_WARN
            4 pgs peering
     monmap e3: 3 mons at {iZbp1isotv99f45cg37lxfZ=10.135.204.114:6789/0,iZbp1isotv99f45cg37lxiZ=10.135.204.132:6789/0,iZbp1isotv99f45cg37lxjZ=10.135.204.107:6789/0}
            election epoch 6, quorum 0,1,2 iZbp1isotv99f45cg37lxjZ,iZbp1isotv99f45cg37lxfZ,iZbp1isotv99f45cg37lxiZ
      fsmap e4: 1/1/1 up {0=iZbp1isotv99f45cg37lxjZ=up:creating}
     osdmap e13: 3 osds: 3 up, 3 in
            flags sortbitwise,require_jewel_osds
      pgmap v20: 80 pgs, 3 pools, 0 bytes data, 0 objects
            100 MB used, 2098 GB / 2098 GB avail
                  64 active+clean
                  12 creating
                   4 creating+peering

在node01部署gateway

docker run -d --name=rgw -p 80:80 \
--restart=always \
-v /etc/ceph:/etc/ceph -v \
/var/lib/ceph/:/var/lib/ceph/ \
ceph/daemon:tag-build-master-jewel-centos-7 rgw

查看ceph状态：

docker exec b3cc55582498 ceph -s

执行结果：

# docker exec b3cc55582498 ceph -s
    cluster 230a93b6-3876-4d69-bffe-c0c054c49653
     health HEALTH_OK
     monmap e3: 3 mons at {iZbp1isotv99f45cg37lxfZ=10.135.204.114:6789/0,iZbp1isotv99f45cg37lxiZ=10.135.204.132:6789/0,iZbp1isotv99f45cg37lxjZ=10.135.204.107:6789/0}
            election epoch 6, quorum 0,1,2 iZbp1isotv99f45cg37lxjZ,iZbp1isotv99f45cg37lxfZ,iZbp1isotv99f45cg37lxiZ
      fsmap e5: 1/1/1 up {0=iZbp1isotv99f45cg37lxjZ=up:active}
     osdmap e14: 3 osds: 3 up, 3 in
            flags sortbitwise,require_jewel_osds
      pgmap v26: 88 pgs, 4 pools, 2068 bytes data, 20 objects
            101 MB used, 2098 GB / 2098 GB avail
                  80 active+clean
                   8 creating

参考

https://judexzhu.gitbooks.io/ceph-docker-deployment/content/Ceph-Docker%20Deployment.html

http://www.jianshu.com/p/f08ed7287416

https://v.qq.com/x/page/h0191o7rpfe.html