Linux Note / 运维笔记

Ceph 简要介绍

Einic Yeo · 8月31日 · 2019年

一、Ceph介绍

​ Ceph是一个可靠地、自动重均衡、自动恢复的分布式存储系统，根据场景划分可以将Ceph分为三大块，分别是对象存储、块设备存储和文件系统服务。在虚拟化领域里，比较常用到的是Ceph的块设备存储，比如在OpenStack项目里，Ceph的块设备存储可以对接OpenStack的cinder后端存储、Glance的镜像存储和虚拟机的数据存储，比较直观的是Ceph集群可以提供一个raw格式的块存储来作为虚拟机实例的硬盘。

​ Ceph相比其它存储的优势点在于它不单单是存储，同时还充分利用了存储节点上的计算能力，在存储每一个数据时，都会通过计算得出该数据存储的位置，尽量将数据分布均衡，同时由于Ceph的良好设计，采用了CRUSH算法、HASH环等方法，使得它不存在传统的单点故障的问题，且随着规模的扩大性能并不会受到影响。

了解什么是块存储/对象存储/文件系统存储？

Ceph 目前提供对象存储（RADOSGW）、块存储RDB以及 CephFS 文件系统这 3 种功能。对于这3种功能介绍，分别如下：
对象存储，也就是通常意义的键值存储，其接口就是简单的GET、PUT、DEL 和其他扩展，代表主要有 Swift 、S3 以及 Gluster 等；
块存储，这种接口通常以 QEMU Driver 或者 Kernel Module 的方式存在，这种接口需要实现 Linux 的 Block Device 的接口或者 QEMU 提供的 Block Driver 接口，如 Sheepdog，AWS 的 EBS，青云的云硬盘和阿里云的盘古系统，还有 Ceph 的 RBD（RBD是Ceph面向块存储的接口）。在常见的存储中 DAS、SAN 提供的也是块存储；
文件存储，通常意义是支持 POSIX 接口，它跟传统的文件系统如 Ext4 是一个类型的，但区别在于分布式存储提供了并行化的能力，如 Ceph 的 CephFS (CephFS是Ceph面向文件存储的接口)，但是有时候又版权声明：本文遵循 CC 4.0 BY-SA 版权协议，若要转载请务必附上原文出处链接及本声明，谢谢合作！会把 GlusterFS ，HDFS 这种非POSIX接口的类文件存储接口归入此类。当然 NFS、NAS也是属于文件系统存储；

1.1 Ceph架构组件

1.2 Ceph核心组件

​ Ceph的核心组件包括Ceph OSD、Ceph Monitor和Ceph MDS。

  Ceph OSD：OSD的英文全称是Object Storage Device，它的主要功能是存储数据、复制数据、平衡数据、恢复数据等，与其它OSD间进行心跳检查等，并将一些变化情况上报给Ceph Monitor。一般情况下一块硬盘对应一个OSD，由OSD来对硬盘存储进行版权声明：本文遵循 CC 4.0 BY-SA 版权协议，若要转载请务必附上原文出处链接及本声明，谢谢合作！管理，当然一个分区也可以成为一个OSD。

  Ceph OSD的架构实现由物理磁盘驱动器、Linux文件系统和Ceph OSD服务组成，对于Ceph OSD Deamon而言，Linux文件系统显性的支持了其拓展性，一般Linux文件系统有好几种，比如有BTRFS、XFS、Ext4等，BTRFS虽然有很多优点特性，但现在还没达到生产环境所需的稳定性，一般比较推荐使用XFS。

  伴随OSD的还有一个概念叫做Journal盘，一般写数据到Ceph集群时，都是先将数据写入到Journal盘中，然后每隔一段时间比如5秒再将Journal盘中的数据刷新到文件系统中。一般为了使读写时延更小，Journal盘都是采用SSD，一般分配10G以上，当然分配多点那是更好，Ceph中引入Journal盘的概念是因为Journal允许Ceph OSD功能很快做小的写操作；一个随机写入首先写入在上一个连续类型的journal，然后刷新到文件系统，这给了文件系统足够的时间来合并写入磁盘，一般情况下使用SSD作为OSD的journal可以有效缓冲突发负载。

  Ceph Monitor：由该英文名字我们可以知道它是一个监视器，负责监视Ceph集群，维护Ceph集群的健康状态，同时维护着Ceph集群中的各种Map图，比如OSD Map、Monitor Map、PG Map和CRUSH Map，这些Map统称为Cluster Map，Cluster Map是RADOS的关键数据结构，管理集群中的所有成员、关系、属性等信息以及数据的分发，比如当用户需要存储数据到Ceph集群时，OSD需要先通过Monitor获取最新的Map图，然后根据Map图和object id等计算出数据最终存储的位置。

  Ceph MDS：全称是Ceph MetaData Server，主要保存的文件系统服务的元数据，但对象存储和块存储设备是不需要使用该服务的。

  查看各种Map的信息可以通过如下命令：ceph osd(mon、pg) dump

二、Ceph部署

环境规划

ceph1 192.168.6.101
ceph2 192.168.6.102

俩台机器都挂俩块盘，一块系统盘，一块osd

配置hosts：

[root@ceph1 ~]# vim /etc/hosts
192.168.6.101   ceph1
192.168.6.102   ceph2

时间同步：

ntpdate ntp1.aliyun.com

设置SSH免密登录：

在ceph1上执行
ssh-keygen
ssh-copy-id ceph1
ssh-copy-id ceph2

配置主机名：

hostnamectl set-hostname ceph1
hostnamectl set-hostname ceph2

2.1 配置ceph.repo

[root@ceph1 ~]# vim /etc/yum.repos.d/ceph.repo
[ceph]
name=Ceph packages for $basearch
baseurl=http://mirrors.aliyun.com/ceph/rpm-luminous/el7/$basearch
enabled=1
gpgcheck=1
priority=1
type=rpm-md
gpgkey=https://mirrors.aliyun.com/ceph/keys/release.asc

[ceph-noarch]
name=Ceph noarch packages
baseurl=http://mirrors.aliyun.com/ceph/rpm-luminous/el7/noarch
enabled=1
gpgcheck=1
priority=1
type=rpm-md
gpgkey=https://mirrors.aliyun.com/ceph/keys/release.asc

[ceph-source]
name=Ceph source packages
baseurl=http://mirrors.aliyun.com/ceph/rpm-luminous/el7/SRPMS
enabled=0
gpgcheck=1
type=rpm-md
gpgkey=https://mirrors.aliyun.com/ceph/keys/release.asc
priority=1

安装ceph-deploy：

yum install -y https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm
yum makecache
yum update -y
yum install -y ceph-deploy python-setuptools

创建一个 Ceph 存储集群，它有一个 Monitor 和两个 OSD 守护进程

mkdir /etc/ceph && cd /etc/ceph
ceph-deploy new ceph1

在ceph.conf中追加以下内容

# 存储集群副本个数
osd_pool_default_size = 2

管理节点和osd节点都需要安装ceph 集群：

ceph-deploy install ceph1 ceph2
版权声明：本文遵循 CC 4.0 BY-SA 版权协议，若要转载请务必附上原文出处链接及本声明，谢谢合作！

配置MON初始化:

ceph-deploy mon create-initial

查看ceph集群状态：

[root@ceph1 ceph]# ceph -s
  cluster:
    id:     cde3244e-89e0-4630-84d5-bf08c0e33b24
    health: HEALTH_OK
 
  services:
    mon: 1 daemons, quorum ceph1
    mgr: no daemons active
    osd: 0 osds: 0 up, 0 in
 
  data:
    pools:   0 pools, 0 pgs
    objects: 0 objects, 0B
    usage:   0B used, 0B / 0B avail
    pgs:

2.2 开启监控模块

ceph-deploy mgr create ceph1

在/etc/ceph/ceph.conf中添加

[mgr]
mgr modules = dashboard

查看集群支持的模块

[root@ceph1 ceph]# ceph mgr dump   
[root@ceph1 ceph]# ceph mgr module enable dashboard   #启用dashboard模块
[root@ceph1 ceph]# ceph mgr dump
{
    "epoch": 3,
    "active_gid": 4110,
    "active_name": "ceph1",
    "active_addr": "192.168.6.101:6800/6619",
    "available": true,
    "standbys": [],
    "modules": [        "balancer",
        "restful",
        "status"
    ],
    "available_modules": [        "balancer",
        "dashboard",
        "influx",
        "localpool",
        "prometheus",
        "restful",
        "selftest",
        "status",
        "zabbix"
    ],
    "services": {}
}
[root@ceph1 ceph]# ceph mgr module enable dashboard
[root@ceph1 ceph]# ceph mgr dump
{
    "epoch": 6,
    "active_gid": 4114,
    "active_name": "ceph1",
    "active_addr": "192.168.6.101:6800/6619",
    "available": true,
    "standbys": [],
    "modules": [        "balancer",
        "dashboard",
        "restful",
        "status"
    ],
    "available_modules": [        "balancer",
        "dashboard",
        "influx",
        "localpool",
        "prometheus",
        "restful",
        "selftest",
        "status",
        "zabbix"
    ],
    "services": {}
}

[root@ceph1 ceph]# ceph -s
  cluster:
    id:     cde3244e-89e0-4630-84d5-bf08c0e33b24
    health: HEALTH_OK
 
  services:
    mon: 1 daemons, quorum ceph1
    mgr: ceph1(active)
    osd: 2 osds: 2 up, 2 in
 
  data:
    pools:   0 pools, 0 pgs
    objects: 0 objects, 0B
    usage:   2.00GiB used, 30.0GiB / 32.0GiB avail
    pgs:

2.3 配置dashboard

[root@ceph-node1 ceph]# ceph config-key put mgr/dashboard/server_addr 192.168.6.101
set mgr/dashboard/server_addr
[root@ceph-node1 ceph]# ceph config-key put mgr/dashboard/server_port 7000
set mgr/dashboard/server_port
[root@ceph1 ~]# netstat -tulnp |grep 7000
tcp        0      0 192.168.6.101:7000      0.0.0.0:*               LISTEN      19836/ceph-mgr

2.4 创建osd

删除磁盘数据

[root@ceph1 ceph]# ceph-deploy disk zap ceph1 /dev/sdb
[root@ceph1 ceph]# ceph-deploy disk zap ceph2 /dev/sdb

创建osd(一共俩个)

[root@ceph1 ceph]# ceph-deploy osd create ceph1 --data /dev/sdb
[root@ceph1 ceph]# ceph-deploy osd create ceph2 --data /dev/sdb

ceph秘钥拷贝（主节点执行）及修改密钥权限
用 c版权声明：本文遵循 CC 4.0 BY-SA 版权协议，若要转载请务必附上原文出处链接及本声明，谢谢合作！eph-deploy 把配置文件和 admin 密钥拷贝到管理节点和 Ceph 节点，这样你每次执行 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

修改密钥权限（所有节点上执行）

[root@ceph1 ceph] # chmod +r /etc/ceph/ceph.client.admin.keyring
[root@ceph2] # chmod +r /etc/ceph/ceph.client.admin.keyring

访问192.168.6.101:7000

danshboard图

三、C版权声明：本文遵循 CC 4.0 BY-SA 版权协议，若要转载请务必附上原文出处链接及本声明，谢谢合作！eph对象存储

3.1 Ceph RGW简介

Ceph RGW(即RADOS Gateway)是Ceph对象存储网关服务，是基于LIBRADOS接口封装实现的FastCGI服务，对外提供存储和管理对象数据的Restful API。 对象存储适用于图片、视频等各类文件的上传下载，可以设置相应的访问权限。目前Ceph RGW兼容常见的对象存储API，例如兼容绝大部分Amazon S3 API，兼容OpenStack Swift API。

3.2 RGW 部署 

[root@ceph1 ceph]# ceph-deploy install --rgw ceph1
[ceph_deploy.conf][DEBUG ] found configuration file at: /root/.cephdeploy.conf
[ceph_deploy.cli][INFO  ] Invoked (2.0.1): /usr/bin/ceph-deploy install --rgw ceph1
[ceph_deploy.cli][INFO  ] ceph-deploy options:
[ceph_deploy.cli][INFO  ]  verbose                       : False
[ceph_deploy.cli][INFO  ]  testing                       : None
[ceph_deploy.cli][INFO  ]  cd_conf                       : <ceph_deploy.conf.cephdeploy.Conf instance at 0x7fa3faca5e60>
[ceph_deploy.cli][INFO  ]  cluster                       : ceph
[ceph_deploy.cli][INFO  ]  dev_commit                    : None
[ceph_deploy.cli][INFO  ]  install_mds                   : False
[ceph_deploy.cli][INFO  ]  stable                        : None
[ceph_deploy.cli][INFO  ]  default_release               : False
[ceph_deploy.cli][INFO  ]  username                      : None
[ceph_deploy.cli][INFO  ]  adjust_repos                  : True
[ceph_deploy.cli][INFO  ]  func                          : <function install at 0x7fa3fbb955f0>
[ceph_deploy.cli][INFO  ]  install_mgr                   : False
[ceph_deploy.cli][INFO  ]  install_all                   : False
[ceph_deploy.cli][INFO  ]  repo                          : False
[ceph_deploy.cli][INFO  ]  host                          : ['ceph1']
[ceph_deploy.cli][INFO  ]  install_rgw                   : True

将配置文件、密钥文件同步到 ceph1： 

[root@ceph1 ceph]# ceph-deploy admin ceph1
[ceph_deploy.conf][DEBUG ] found configuration file at: /root/.cephdeploy.conf
[ceph_deploy.cli][INFO  ] Invoked (2.0.1): /usr/bin/ceph-deploy admin ceph1
[ceph_deploy.cli][INFO  ] ceph-deploy options:
[ceph_deploy.cli][INFO  ]  username                      : None
[ceph_deploy.cli][INFO  ]  verbose                       : False
[ceph_deploy.cli][INFO  ]  overwrite_conf                : False
[ceph_deploy.cli][INFO  ]  quiet                         : False
[ceph_deploy.cli][INFO  ]  cd_conf                       : <ceph_deploy.conf.cephdeploy.Conf instance at 0x7fe0e152d3b0>
[ceph_deploy.cli][INFO  ]  cluster                       : ceph
[ceph_deploy.cli][INFO  ]  client                        : ['ceph1']
[ceph_deploy.cli][INFO  ]  func                          : <function admin at 0x7fe0e1dc0230>
[ceph_deploy.cli][INFO  ]  ceph_conf                     : None
[ceph_deploy.cli][INFO  ]  default_release               : False
[ceph_deploy.admin][DEBUG ] Pushing admin keys and conf to ceph1
[ceph1][DEBUG ] connected to host: ceph1 
[ceph1][DEBUG ] detect platform information from remote host
[ceph1][DEBUG ] detect machine type
[ceph1][DEBUG ] write cluster configuration to /etc/ceph/{cluster}.conf

启动一个RGW服务 ： 

先将ceph.conf加一个参数配置
[root@ceph1 ceph]# vim ceph.conf 
[global]
fsid = cde3244e-89e0-4630-84d5-bf08c0e33b24
mon_initial_members = ceph1
mon_host = 192.168.6.101
auth_cluster_required = cephx
auth_service_required = cephx
auth_client_required = cephx
osd_pool_default_size = 2
[mgr]
mgr modules = dashboard
[mon]
mon allow pool delete = true   ####有这个配置，生成的pool才可以被删除



[root@ceph1 ceph]#  ceph-deploy rgw create ceph1
[ceph_deploy.conf][DEBUG ] found configuration file at: /root/.cephdeploy.conf
[ceph_deploy.cli][INFO  ] Invoked (2.0.1): /usr/bin/ceph-deploy rgw create ceph1
[ceph_deploy.cli][INFO  ] ceph-deploy options:
[ceph_deploy.cli][INFO  ]  username                      : None
[ceph_deploy.cli][INFO  ]  verbose                       : False
[ceph_deploy.cli][INFO  ]  rgw                           : [('ceph1', 'rgw.ceph1')]
[ceph_deploy.cli][INFO  ]  overwrite_conf                : False
[ceph_deploy.cli][INFO  ]  subcommand                    : create
[ceph_deploy.cli][INFO  ]  quiet                         : False
[ceph_deploy.cli][INFO  ]  cd_conf                       : <ceph_deploy.conf.cephdeploy.Conf instance at 0x7fda85404ab8>
[ceph_deploy.cli][INFO  ]  cluster                       : ceph
[ceph_deploy.cli][INFO  ]  func                          : <function rgw at 0x7fda85a53050>
[ceph_deploy.cli][INFO  ]  ceph_conf                     : None
[ceph_deploy.cli][INFO  ]  default_release               : False
[ceph_deploy.rgw][DEBUG ] Deploying rgw, cluster ceph hosts ceph1:rgw.ceph1
[ceph1][DEBUG ] connected to host: ceph1 
[ceph1][DEBUG ] detect platform information from remote host
[ceph1][DEBUG ] detect machine type
[ceph_deploy.rgw][INFO  ] Distro info: CentOS Linux 7.6.1810 Core
[ceph_deploy.rgw][DEBUG ] remote host will use systemd
[ceph_deploy.rgw][DEBUG ] deploying rgw bootstrap to ceph1
[ceph1][DEBUG ] write cluster configuration to /etc/ceph/{cluster}.conf
[ceph1][DEBUG ] create path recursively if it doesn't exist
[ceph1][INFO  ] Running command: ceph --cluster ceph --name client.bootstrap-rgw --keyring /var/lib/ceph/bootstrap-rgw/ceph.keyring auth get-or-create client.rgw.ceph1 osd allow rwx mon allow rw -o /var/lib/ceph/radosgw/ceph-rgw.ceph1/keyring
[ceph1][INFO  ] Running command: systemctl enable [email protected]
[ceph1][INFO  ] Running command: systemctl start [email protected]
[ceph1][INFO  ] Running command: systemctl enable ceph.target
[ceph_deploy.rgw][INFO  ] The Ceph Object Gateway (RGW) is now running on host ceph1 and default port 7480

验证：
[root@ceph1 ceph]# systemctl status [email protected]
● [email protected] - Ceph rados gateway
   Loaded: loaded (/usr/lib/systemd/system/[email protected]; enabled; vendor preset: disabled)
   Active: active (running) since Thu 2019-08-31 09:03:24 CST; 9s ago
 Main PID: 21057 (radosgw)
   CGroup: /system.slice/system-ceph\x2dradosgw.slice/[email protected]
           └─21057 /usr/bin/radosgw -f --cluster ceph --name client.rgw.ceph1 --setuser ceph --setgroup ceph

Jul 11 09:03:24 ceph1 systemd[1]: Started Ceph rados gateway.
[root@ceph1 ceph]# ceph -s
  cluster:
    id:     cde3244e-89e0-4630-84d5-bf08c0e33b24
    health: HEALTH_OK
 
  services:
    mon: 1 daemons, quorum ceph1
    mgr: ceph1(active)
    osd: 2 osds: 2 up, 2 in
    rgw: 1 daemon active
 
  data:
    pools:   4 pools, 32 pgs
    objects: 187 objects, 1.09KiB
    usage:   2.01GiB used, 30.0GiB / 32.0GiB avail
    pgs:     32 active+clean
danshboard图

3.3 s3 用户管理

创建用户 ：

[root@ceph1 ceph]# radosgw-admin user create --uid="radosgw" --display-name="First User"
{
    "user_id": "radosgw",
    "display_name": "First User",
    "email": "",
    "suspended": 0,
    "max_buckets": 1000,
    "auid": 0,
    "subusers": [],
    "keys": [
        {
            "user": "radosgw",
            "access_key": "CQE7E6ZDVA74KVJ0077A",
            "secret_key": "wnRy76RM2s85ozWvKwTBezrBU0RwcCTQJf1HFAM8"
        }
    ],
    "swift_keys": [],
    "caps": [],
    "op_mask": "read, write, delete",
    "default_placement": "",
    "placement_tags": [],
    "bucket_quota": {
        "enabled": false,
        "check_on_raw": false,
        "max_size": -1,
        "max_size_kb": 0,
        "max_objects": -1
    },
    "user_quota": {
        "enabled": false,
        "check_on_raw": false,
        "max_size": -1,
        "max_size_kb": 0,
        "max_objects": -1
    },
    "temp_url_keys": [],
    "type": "rgw"
}

这个是后续需要的账户信息 

"user": "radosgw",
            "access_key": "CQE7E6ZDVA74KVJ0077A",
            "secret_key": "wnRy76RM2s85ozWvKwTBezrBU0RwcCTQJf1HFAM8"

授权用户，允许 radosgw 读写 users 信息： 

[root@ceph1 ceph]# radosgw-admin caps add --uid=radosgw --caps="users=*"
{
    "user_id": "radosgw",
    "display_name": "First User",
    "email": "",
    "suspended": 0,
    "max_buckets": 1000,
    "auid": 0,
    "subusers": [],
    "keys": [
        {
            "user": "radosgw",
            "access_key": "CQE7E6ZDVA74KVJ0077A",
            "secret_key": "wnRy76RM2s85ozWvKwTBezrBU0RwcCTQJf1HFAM8"
        }
    ],
    "swift_keys": [],
    "caps": [
        {
            "type": "users",
            "perm": "*"
        }
    ],
    "op_mask": "read, write, delete",
    "default_placement": "",
    "placement_tags": [],
    "bucket_quota": {
        "enabled": false,
        "check_on_raw": false,
        "max_size": -1,
        "max_size_kb": 0,
        "max_objects": -1
    },
    "user_quota": {
        "enabled": false,
        "check_on_raw": false,
        "max_size": -1,
        "max_size_kb": 0,
        "max_objects": -1
    },
    "temp_url_keys": [],
    "type": "rgw"
}

允许 radosgw 读写所有的usage信息： 

[root@ceph1 ceph]#  radosgw-admin caps add --uid=radosgw --caps="usage=read,write"
{
    "user_id": "radosgw",
    "display_name": "First User",
    "email": "",
    "suspended": 0,
    "max_buckets": 1000,
    "auid": 0,
    "subusers": [],
    "keys": [
        {
            "user": "radosgw",
            "access_key": "CQE7E6ZDVA74KVJ0077A",
            "secret_key": "wnRy76RM2s85ozWvKwTBezrBU0RwcCTQJf1HFAM8"
        }
    ],
    "swift_keys": [],
    "caps": [
        {
            "type": "usage",
            "perm": "*"
        },
        {
            "type": "users",
            "perm": "*"
        }
    ],
    "op_mask": "read, write, delete",
    "default_placement": "",
    "placement_tags": [],
    "bucket_quota": {
        "enabled": false,
        "check_on_raw": false,
        "max_size": -1,
        "max_size_kb": 0,
        "max_objects": -1
    },
    "user_quota": {
        "enabled": false,
        "check_on_raw": false,
        "max_size": -1,
        "max_size_kb": 0,
        "max_objects": -1
    },
    "temp_url_keys": [],
    "type": "rgw"
}

创建子用户，做为后面 swift 客户端访问时使用： 

[root@ceph1 ceph]#  radosgw-admin subuser create --uid=radosgw --subuser=radosgw:swift --access=full
{
    "user_id": "radosgw",
    "display_name": "First User",
    "email": "",
    "suspended": 0,
    "max_buckets": 1000,
    "auid": 0,
    "subusers": [
        {
            "id": "radosgw:swift",
            "permissions": "full-control"
        }
    ],
    "keys": [
        {
            "user": "radosgw",
            "access_key": "CQE7E6ZDVA74KVJ0077A",
            "secret_key": "wnRy76RM2s85ozWvKwTBezrBU0RwcCTQJf1HFAM8"
        }
    ],
    "swift_keys": [
        {
            "user": "radosgw:swift",
            "secret_key": "A3GDj2yjkGJahkCM6YJS4QKQlGz2zd65GXvCkiwV"
        }
    ],
    "caps": [
        {
            "type": "usage",
            "perm": "*"
        },
        {
            "type": "users",
            "perm": "*"
        }
    ],
    "op_mask": "read, write, delete",
    "default_placement": "",
    "placement_tags": [],
    "bucket_quota": {
        "enabled": false,
        "check_on_raw": false,
        "max_size": -1,
        "max_size_kb": 0,
        "max_objects": -1
    },
    "user_quota": {
        "enabled": false,
        "check_on_raw": false,
        "max_size": -1,
        "max_size_kb": 0,
        "max_objects": -1
    },
    "temp_url_keys": [],
    "type": "rgw"
}

创建密钥 ： 

[root@ceph1 ceph]# radosgw-admin key create --subuser=radosgw:swift --key-type=swift --gen-secret
{
    "user_id": "radosgw",
    "display_name": "First User",
    "email": "",
    "suspended": 0,
    "max_buckets": 1000,
    "auid": 0,
    "subusers": [
        {
            "id": "radosgw:swift",
            "permissions": "full-control"
        }
    ],
    "keys": [
        {
            "user": "radosgw",
            "access_key": "CQE7E6ZDVA74KVJ0077A",
            "secret_key": "wnRy76RM2s85ozWvKwTBezrBU0RwcCTQJf1HFAM8"
        }
    ],
    "swift_keys": [
        {
            "user": "radosgw:swift",
            "secret_key": "CZ2F5lqOfdmTCLSPyzqcw5pPNeetruAL4HjHkiAh"
        }
    ],
    "caps": [
        {
            "type": "usage",
            "perm": "*"
        },
        {
            "type": "users",
            "perm": "*"
        }
    ],
    "op_mask": "read, write, delete",
    "default_placement": "",
    "placement_tags": [],
    "bucket_quota": {
        "enabled": false,
        "check_on_raw": false,
        "max_size": -1,
        "max_size_kb": 0,
        "max_objects": -1
    },
    "user_quota": {
        "enabled": false,
        "check_on_raw": false,
        "max_size": -1,
        "max_size_kb": 0,
        "max_objects": -1
    },
    "temp_url_keys": [],
    "type": "rgw"
}

3.4 s3 安装与配置

[root@ceph1 ceph]# yum -y install s3cmd.noarch
[root@ceph1 ceph]# s3cmd --configure
New settings:
  Access Key: CQE7E6ZDVA74KVJ0077A
  Secret Key: wnRy76RM2s85ozWvKwTBezrBU0RwcCTQJf1HFAM8
  Default Region: US
  S3 Endpoint: 192.168.6.101:7480
  DNS-style bucket+hostname:port template for accessing a bucket: %(bucket)s.192.168.6.101:7480 bucket
  Encryption password: 123456
  Path to GPG program: /usr/bin/gpg
  Use HTTPS protocol: False
  HTTP Proxy server name: 
  HTTP Proxy server port: 0

Test access with supplied credentials? [Y/n] y
Please wait, attempting to list all buckets...
Success. Your access key and secret key worked fine :-)

Now verifying that encryption works...
Success. Encryption and decryption worked fine :-)

Save settings? [y/N] y
Configuration saved to '/root/.s3cfg'

格式是这样： 

Default Region [US]:                        #这里一定不要修改，否则后面会报错
S3 Endpoint [s3.amazonaws.com]: 192.168.6.101:7480
DNS-style bucket+hostname:port template for accessing a bucket [%(bucket)s.s3.amazonaws.com]: %(bucket)s.192.168.20.148:7480 bucket     #相当于百度网盘的创建文件夹，这里是固定格式
Path to GPG program [/usr/bin/gpg]:                 #保持默认
Use HTTPS protocol [Yes]: no                    #这里写 no ，因为没有提供 https 端口
HTTP Proxy server name:                       #这里不用写，因为没有代理
Test access with supplied credentials? [Y/n] y
Save settings? [y/N] y

3.5 bucket 创建 

root@ceph1 ~]# s3cmd mb s3://cephdir
[root@ceph1 ~]# s3cmd put /etc/hosts s3://ceph_dir
upload: '/etc/hosts' -> 's3://ceph_dir/hosts'  [1 of 1]
 200 of 200   100% in    1s   133.14 B/s  done
 
 [root@ceph1 ~]# s3cmd ls s3://ceph_dir
2019-08-31 09:41       200   s3://ceph_dir/hosts

s3 的测试脚本： 

[root@ceph-f ~]# yum -y install python-boto
[root@ceph-f ~]# vim s3test.py
import boto.s3.connection
access_key = 'N6ALEK0KS0ISYCIM5JBG'
secret_key = 'qK9hrpX2uwna4elPP1VsuErmAHBw3So40fE2K4yM'
conn = boto.connect_s3(
         aws_access_key_id=access_key,         
         aws_secret_access_key=secret_key,         
         host='ceph1', port=7480,         
         is_secure=False, calling_format=boto.s3.connection.OrdinaryCallingFormat(),        
         )
bucket = conn.create_bucket('xxx_yyy')
for bucket in conn.get_all_buckets():
     print "{name} {created}".format(         
     name=bucket.name,         
     created=bucket.creation_date,     
     )

在使用时，请替换自己的 access_key、secret_key、主机名和端口

3.6 swift 接口测试

yum install python-setuptools
easy_install pip
pip install --upgrade setuptools
pip install --upgrade python-swiftclient

命令行访问 ： 

[root@ceph1 ~]# swift -A http://192.168.6.101:7480/auth/v1.0 -U radosgw:swift -K CZ2F5lqOfdmTCLSPyzqcw5pPNeetruAL4HjHkiAh list
ceph_dir


swift -A http://192.168.6.101:7480/auth/v1.0 -U radosgw:swift -K CZ2F5lqOfdmTCLSPyzqcw5pPNeetruAL4HjHkiAh post sy-yt

swift -A http://192.168.6.101:7480/auth/v1.0 -U radosgw:swift -K CZ2F5lqOfdmTCLSPyzqcw5pPNeetruAL4HjHkiAh post sy_yt



[root@ceph1 ~]# swift -A http://192.168.6.101:7480/auth/v1.0 -U radosgw:swift -K CZ2F5lqOfdmTCLSPyzqcw5pPNeetruAL4HjHkiAh list
ceph_dir
sy-yt
sy_yt

这里提供 swift 的测试脚本： 

[root@ceph-f ~]# vim swift.py
import swiftclient
user = 'radosgw:swift'
key = 'CZ2F5lqOfdmTCLSPyzqcw5pPNeetruAL4HjHkiAh'

conn = swiftclient.Connection(
         user=user,         
         key=key,         
         authurl='http://192.168.6.101:7480/auth/v1.0',
         )
for container in conn.get_account()[1]:
         print container['name']

在使用时，请替换自己的 access_key、secret_key、authurl

四、Ceph集群部署

4.1 集群架构

ceph1 192.168.10.186 admin、mon、mgr、osd、rgw
ceph2 192.168.10.187 mon、mgr、osd、rgw
ceph3 192.168.10.188 mon、mgr、osd、rgw

环境初始化

[root@10dot186 ~]# vim /etc/hosts
192.168.10.186   ceph1
192.168.10.187   ceph2
192.168.10.188   ceph3

hostnamectl set-hostname ceph1
hostnamectl set-hostname ceph2
hostnamectl set-hostname ceph3

ntpdate ntp1.aliyun.com



ssh-keygen
ssh-copy-id ceph1
ssh-copy-id ceph2
ssh-copy-id ceph3


[root@ceph1 ~]# vim /etc/yum.repos.d/ceph.repo
[ceph]
name=Ceph packages for $basearch
baseurl=http://mirrors.aliyun.com/ceph/rpm-luminous/el7/$basearch
enabled=1
gpgcheck=1
priority=1
type=rpm-md
gpgkey=https://mirrors.aliyun.com/ceph/keys/release.asc

[ceph-noarch]
name=Ceph noarch packages
baseurl=http://mirrors.aliyun.com/ceph/rpm-luminous/el7/noarch
enabled=1
gpgcheck=1
priority=1
type=rpm-md
gpgkey=https://mirrors.aliyun.com/ceph/keys/release.asc

[ceph-source]
name=Ceph source packages
baseurl=http://mirrors.aliyun.com/ceph/rpm-luminous/el7/SRPMS
enabled=0
gpgcheck=1
type=rpm-md
gpgkey=https://mirrors.aliyun.com/ceph/keys/release.asc
priority=1




yum install -y https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm
yum makecache
yum update -y
yum install -y ceph-deploy




mkdir /etc/ceph && cd /etc/ceph
ceph-deploy new ceph1  ceph2 ceph3


yum install -y python-setuptools

在配置文件中增加：
osd_pool_default_size = 3
[mgr]
mgr modules = dashboard
[mon]
mon allow pool delete = true

4.2 mon部署

ceph-deploy install ceph1 ceph2 ceph3

ceph-deploy mon create-initial



[root@ceph1 ceph]# ceph -s
  cluster:
    id:     fcb2fa5e-481a-4494-9a27-374048f37113
    health: HEALTH_OK
 
  services:
    mon: 3 daemons, quorum ceph1,ceph2,ceph3
    mgr: no daemons active
    osd: 0 osds: 0 up, 0 in
 
  data:
    pools:   0 pools, 0 pgs
    objects: 0 objects, 0B
    usage:   0B used, 0B / 0B avail
    pgs:

4.3 mgr部署

ceph-deploy mgr create ceph1 ceph2 ceph3


[root@ceph1 ceph]# ceph -s
  cluster:
    id:     fcb2fa5e-481a-4494-9a27-374048f37113
    health: HEALTH_OK
 
  services:
    mon: 3 daemons, quorum ceph1,ceph2,ceph3
    mgr: ceph1(active), standbys: ceph2, ceph3
    osd: 0 osds: 0 up, 0 in
 
  data:
    pools:   0 pools, 0 pgs
    objects: 0 objects, 0B
    usage:   0B used, 0B / 0B avail
    pgs: 
	
	
	
[root@ceph1 ceph]# ceph mgr dump
{
    "epoch": 4,
    "active_gid": 4122,
    "active_name": "ceph1",
    "active_addr": "192.168.10.186:6800/22316",
    "available": true,
    "standbys": [
        {
            "gid": 4129,
            "name": "ceph2",
            "available_modules": [
                "balancer",
                "dashboard",
                "influx",
                "localpool",
                "prometheus",
                "restful",
                "selftest",
                "status",
                "zabbix"
            ]
        },
        {
            "gid": 4132,
            "name": "ceph3",
            "available_modules": [
                "balancer",
                "dashboard",
                "influx",
                "localpool",
                "prometheus",
                "restful",
                "selftest",
                "status",
                "zabbix"
            ]
        }
    ],
    "modules": [
        "balancer",
        "restful",
        "status"
    ],
    "available_modules": [
        "balancer",
        "dashboard",
        "influx",
        "localpool",
        "prometheus",
        "restful",
        "selftest",
        "status",
        "zabbix"
    ],
    "services": {}
}
[root@ceph1 ceph]# ceph mgr module enable dashboard
[root@ceph1 ceph]# ceph mgr dump
{
    "epoch": 7,
    "active_gid": 4139,
    "active_name": "ceph1",
    "active_addr": "192.168.10.186:6800/22316",
    "available": true,
    "standbys": [
        {
            "gid": 4136,
            "name": "ceph3",
            "available_modules": [
                "balancer",
                "dashboard",
                "influx",
                "localpool",
                "prometheus",
                "restful",
                "selftest",
                "status",
                "zabbix"
            ]
        },
        {
            "gid": 4141,
            "name": "ceph2",
            "available_modules": [
                "balancer",
                "dashboard",
                "influx",
                "localpool",
                "prometheus",
                "restful",
                "selftest",
                "status",
                "zabbix"
            ]
        }
    ],
    "modules": [
        "balancer",
        "dashboard",
        "restful",
        "status"
    ],
    "available_modules": [
        "balancer",
        "dashboard",
        "influx",
        "localpool",
        "prometheus",
        "restful",
        "selftest",
        "status",
        "zabbix"
    ],
    "services": {}
}





[root@ceph1 ceph]# ceph config-key put mgr/dashboard/server_addr 192.168.6.101
set mgr/dashboard/server_addr
[root@ceph1 ceph]# ceph config-key put mgr/dashboard/server_port 7000
set mgr/dashboard/server_port
[root@ceph1 ~]# netstat -tulnp |grep 7000
tcp        0      0 192.168.6.101:7000      0.0.0.0:*               LISTEN      19836/ceph-mgr
danshboard图

4.4 OSD部署

每台机器做逻辑卷
[root@ceph1 ceph]# pvcreate /dev/sdb
  Physical volume "/dev/sdb" successfully created.

[root@ceph1 ceph]# vgcreate data_vg1 /dev/sdb
  Volume group "data_vg1" successfully created
  
[root@ceph1 ceph]# lvcreate -n data_lv1 -L 99g data_vg1   
  Logical volume "data_lv1" created.
  



ceph-deploy osd create ceph1 --data data_vg1/data_lv1
ceph-deploy osd create ceph2 --data data_vg1/data_lv1
ceph-deploy osd create ceph3 --data data_vg1/data_lv1

[root@ceph1 ceph]# ceph -s
  cluster:
    id:     fcb2fa5e-481a-4494-9a27-374048f37113
    health: HEALTH_OK
 
  services:
    mon: 3 daemons, quorum ceph1,ceph2,ceph3
    mgr: ceph1(active), standbys: ceph3, ceph2
    osd: 3 osds: 3 up, 3 in
 
  data:
    pools:   0 pools, 0 pgs
    objects: 0 objects, 0B
    usage:   3.01GiB used, 294GiB / 297GiB avail
    pgs:
danshboard图

4.5 rgw部署

ceph-deploy install --rgw ceph1 ceph2 ceph3

ceph-deploy admin ceph1 ceph2 ceph3

ceph-deploy rgw create ceph1 ceph2 ceph3

[root@ceph1 ceph]# ceph -s
  cluster:
    id:     fcb2fa5e-481a-4494-9a27-374048f37113
    health: HEALTH_OK
 
  services:
    mon: 3 daemons, quorum ceph1,ceph2,ceph3
    mgr: ceph1(active), standbys: ceph3, ceph2
    osd: 3 osds: 3 up, 3 in
    rgw: 3 daemons active
 
  data:
    pools:   4 pools, 32 pgs
    objects: 191 objects, 3.08KiB
    usage:   3.01GiB used, 294GiB / 297GiB avail
    pgs:     32 active+clean
danshboard图

4.6 NGINX 配置代理

[root@ceph1 conf.d]# cat cephcloud.infvie.org.conf 
        upstream cephcloud.infvie.org  {
        server  192.168.10.186:7480;
        server  192.168.10.187:7480;
        server  192.168.10.188:7480;        
        }
        server {
        listen       80;
        server_name  cephcloud.infvie.org;
        location / {
                        proxy_intercept_errors on;
                        access_log /var/log/nginx/cephcloud_log;
                        proxy_pass http://cephcloud.infvie.org;
                        proxy_set_header X-Real-IP $remote_addr;
                        proxy_set_header Host $host;
                        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
                        proxy_set_header Request_Uri $request_uri;

                }
        }

4.7 s3&swift 配置 

New settings:
  Access Key: M954JYYAOBES65B7UNEZ
  Secret Key: 11MZu3N9vB4S4C4N8U2Ywgkhxro3Xi6K9HPyRQ9v
  Default Region: US
  S3 Endpoint: cephcloud.dev.goago.cn
  DNS-style bucket+hostname:port template for accessing a bucket: %(bucket)s.cephcloud.dev.goago.cn bucket
  Encryption password: 123456
  Path to GPG program: /usr/bin/gpg
  Use HTTPS protocol: False
  HTTP Proxy server name: 
  HTTP Proxy server port: 0

参考文献

https://www.cnblogs.com/yangxiaoyi/p/7795274.html

https://github.com/ceph/ceph

https://ceph.io/install/

https://www.cnblogs.com/Honeycomb/p/9768380.html

相关文章

ceph 存储节点系统盘损坏集群恢复

2月14日 · 2022年

Google SRE 二十年的经验教训

11月20日 · 2023年

虚机网格(istio)管理实战篇

11月17日 · 2023年

云原生混沌工程平台 – ChaosMeta

11月9日 · 2023年
0 条回应
© 2017~ Infvie Envoy. All Rights Reserved.
IPV6 | GITHUB