手动部署集群¶

编译构建¶

使用如下命令同时构建server，client及相关的依赖：

$ git clone http://github.com/chubaofs/chubaofs.git
$ cd chubaofs
$ make build

如果构建成功，将在`build/bin` 目录中生成可执行文件`cfs-server`和`cfs-client`。

集群部署¶

启动资源管理节点¶

nohup ./cfs-server -c master.json &

示例 master.json ：注意：master服务最少应该启动3个节点实例

{
  "role": "master",
  "ip": "10.196.59.198",
  "listen": "17010",
  "prof":"17020",
  "id":"1",
  "peers": 1:10.196.59.198:17010,2:10.196.59.199:17010,3:10.196.59.200:17010",
  "retainLogs":"20000",
  "logDir": "/cfs/master/log",
  "logLevel":"info",
  "walDir":"/cfs/master/data/wal",
  "storeDir":"/cfs/master/data/store",
  "consulAddr": "http://consul.prometheus-cfs.local",
  "exporterPort": 9500,
  "clusterName":"chubaofs01",
  "metaNodeReservedMem": "1073741824"
}

详细配置参数请参考资源管理节点。

启动元数据节点¶

nohup ./cfs-server -c metanode.json &

示例 meta.json ：注意：metanode服务最少应该启动3个节点实例

{
    "role": "metanode",
    "listen": "17210",
    "prof": "17220",
    "logLevel": "info",
    "metadataDir": "/cfs/metanode/data/meta",
    "logDir": "/cfs/metanode/log",
    "raftDir": "/cfs/metanode/data/raft",
    "raftHeartbeatPort": "17230",
    "raftReplicaPort": "17240",
    "totalMem":  "8589934592",
    "consulAddr": "http://consul.prometheus-cfs.local",
    "exporterPort": 9501,
    "masterAddr": [
        "10.196.59.198:17010",
        "10.196.59.199:17010",
        "10.196.59.200:17010"
    ]
}

启动元数据节点¶

nohup ./cfs-server -c metanode.json &

示例 meta.json ：注意：metanode服务最少应该启动3个节点实例

{
    "role": "metanode",
    "listen": "17210",
    "prof": "17220",
    "logLevel": "info",
    "metadataDir": "/cfs/metanode/data/meta",
    "logDir": "/cfs/metanode/log",
    "raftDir": "/cfs/metanode/data/raft",
    "raftHeartbeatPort": "17230",
    "raftReplicaPort": "17240",
    "totalMem":  "8589934592",
    "consulAddr": "http://consul.prometheus-cfs.local",
    "exporterPort": 9501,
    "masterAddr": [
        "10.196.59.198:17010",
        "10.196.59.199:17010",
        "10.196.59.200:17010"
    ]
}

启动动 ObjectNode¶

nohup ./cfs-server -c objectnode.json &

示例 objectnode.json 内容如下

{
    "role": "objectnode",
    "domains": [
        "object.cfs.local"
    ],
    "listen": 17410,
    "masterAddr": [
       "10.196.59.198:17010",
       "10.196.59.199:17010",
       "10.196.59.200:17010"
    ],
    "logLevel": "info",
    "logDir": "/cfs/Logs/objectnode"
}

配置文件的详细信息 objectnode.json, 请参阅对象存储(ObjectNode).

启动管理平台（非必须）¶

nohup ./cfs-server -c console.json &

示例 console.json 内容如下

{
    "role": "console",
    "logDir": "/cfs/log/",
    "logLevel": "debug",
    "listen": "80",
    "masterAddr": [
        "192.168.0.11:17010",
        "192.168.0.12:17010",
        "192.168.0.13:17010"
    ],
    "objectNodeDomain": "object.chubao.io",
    "master_instance": "192.168.0.11:9066",
    "monitor_addr": "http://192.168.0.102:9090",
    "dashboard_addr": "http://192.168.0.103",
    "monitor_app": "cfs",
    "monitor_cluster": "cfs"
}

配置文件的详细信息 console.json, 请参阅 Console.

详细配置参数请参考元数据节点.

启动数据节点¶

准备数据目录

推荐使用单独磁盘作为数据目录，配置多块磁盘能够达到更高的性能。

磁盘准备
1.1 查看机器磁盘信息，选择给ChubaoFS使用的磁盘
fdisk -l
1.2 格式化磁盘，建议格式化为XFS
mkfs.xfs -f /dev/sdx
1.3 创建挂载目录
mkdir /data0
1.4 挂载磁盘
mount /dev/sdx /data0

启动数据节点

nohup ./cfs-server -c datanode.json &

示例 datanode.json :注意：datanode服务最少应该启动4个节点实例

{
  "role": "datanode",
  "listen": "17310",
  "prof": "17320",
  "logDir": "/cfs/datanode/log",
  "logLevel": "info",
  "raftHeartbeat": "17330",
  "raftReplica": "17340",
  "raftDir":"/cfs/datanode/log",
  "consulAddr": "http://consul.prometheus-cfs.local",
  "exporterPort": 9502,
  "masterAddr": [
     "10.196.59.198:17010",
     "10.196.59.199:17010",
     "10.196.59.200:17010"
  ],
  "disks": [
     "/data0:10737418240",
     "/data1:10737418240"
 ]
}

详细配置参数请参考数据节点.

启动对象管理节点¶

nohup ./cfs-server -c objectnode.json &

示例 objectnode.json is 如下：

{
    "role": "objectnode",
    "domains": [
        "object.cfs.local"
    ],
    "listen": 17410,
    "masterAddr": [
       "10.196.59.198:17010",
       "10.196.59.199:17010",
       "10.196.59.200:17010"
    ],
    "logLevel": "info",
    "logDir": "/cfs/Logs/objectnode"
}

关于 object.json 的更多详细配置请参考对象存储(ObjectNode).

创建Volume卷¶

curl -v "http://10.196.59.198:17010/admin/createVol?name=test&capacity=10000&owner=cfs"

如果执行性能测试，请调用相应的API，创建足够多的数据分片（data partition）,如果集群中有8块磁盘，那么需要创建80个datapartition

挂载客户端¶

运行 modprobe fuse 插入FUSE内核模块。
运行 yum install -y fuse 安装libfuse。

运行 nohup client -c fuse.json & 启动客户端。

样例 fuse.json ,

{
  "mountPoint": "/cfs/mountpoint",
  "volName": "ltptest",
  "owner": "ltptest",
  "masterAddr": "10.196.59.198:17010,10.196.59.199:17010,10.196.59.200:17010",
  "logDir": "/cfs/client/log",
  "profPort": "17510",
  "exporterPort": "9504",
  "logLevel": "info"
}

详细配置参数请参考客户端.

用户可以使用不同的挂载点在同一台机器上同时启动多个客户端

升级注意事项¶

集群数据节点和元数据节点升级前，请先禁止集群自动为卷扩容数据分片.

冻结集群

curl -v "http://10.196.59.198:17010/cluster/freeze?enable=true"

升级节点
开启自动扩容数据分片

curl -v "http://10.196.59.198:17010/cluster/freeze?enable=false"

注：升级节点时不能修改各节点配置文件的端口。