Zookeeper学习之路（三）shell操作

HK123COM 2019-02-14

展开全文

正文

Zookeeper的shell操作

Zookeeper命令工具

在启动Zookeeper服务之后，输入以下命令，连接到Zookeeper服务：

[hadoop@hadoop1 ~]$ zkCli.sh -server hadoop2:2181

View Code

连接成功之后，系统会输出Zookeeper的相关环境及配置信息，并在屏幕输出“welcome to Zookeeper！”等信息。

输入help之后，屏幕会输出可用的Zookeeper命令，如下图所示

使用Zookeeper命令的简单操作步骤

(1) 使用ls命令查看当前Zookeeper中所包含的内容：ls /

[zk: hadoop2:2181(CONNECTED) 1] ls /
[zookeeper]
[zk: hadoop2:2181(CONNECTED) 2]

(2) 创建一个新的Znode节点"aa"，以及和它相关字符，执行命令：create /aa "my first zk"，默认是不带编号的

[zk: hadoop2:2181(CONNECTED) 2] create /aa "my first zk"
Created /aa
[zk: hadoop2:2181(CONNECTED) 3]

　　创建带编号的持久性节点"bb"，

[zk: localhost:2181(CONNECTED) 1] create -s /bb "bb"
Created /bb0000000001
[zk: localhost:2181(CONNECTED) 2]

　　创建不带编号的临时节点"cc"

[zk: localhost:2181(CONNECTED) 2] create -e /cc "cc"
Created /cc
[zk: localhost:2181(CONNECTED) 3]

　　创建带编号的临时节点"dd"

[zk: localhost:2181(CONNECTED) 3] create -s -e /dd "dd"
Created /dd0000000003
[zk: localhost:2181(CONNECTED) 4]

(3) 再次使用ls命令来查看现在Zookeeper的中所包含的内容：ls /

[zk: localhost:2181(CONNECTED) 4] ls /
[cc, dd0000000003, zookeeper, bb0000000001]
[zk: localhost:2181(CONNECTED) 5]

此时看到，aa节点已经被创建。

关闭本次连接回话session，再重新打开一个连接

[zk: localhost:2181(CONNECTED) 5] close
2018-03-22 13:03:29,137 [myid:] - INFO  [main:ZooKeeper@684] - Session: 0x1624c10e8d90000 closed
[zk: localhost:2181(CLOSED) 6] 2018-03-22 13:03:29,139 [myid:] - INFO  [main-EventThread:ClientCnxn$EventThread@519] - EventThread shut down for session: 0x1624c10e8d90000

[zk: localhost:2181(CLOSED) 6] ls /
Not connected
[zk: localhost:2181(CLOSED) 7] connect hadoop1:2181

重新查看，临时节点已经随着上一次的会话关闭自动删除了

[zk: hadoop1:2181(CONNECTED) 8] ls /
[zookeeper, bb0000000001]
[zk: hadoop1:2181(CONNECTED) 9]

(4) 使用get命令来确认第二步中所创建的Znode是否包含我们创建的字符串，执行命令：get /aa

[zk: hadoop2:2181(CONNECTED) 4] get /aa
my first zk
cZxid = 0x100000002
ctime = Wed Mar 21 20:01:02 CST 2018
mZxid = 0x100000002
mtime = Wed Mar 21 20:01:02 CST 2018
pZxid = 0x100000002
cversion = 0
dataVersion = 0
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 11
numChildren = 0
[zk: hadoop2:2181(CONNECTED) 5]

(5) 接下来通过set命令来对zk所关联的字符串进行设置，执行命令：set /aa haha123

[zk: hadoop2:2181(CONNECTED) 6] set /aa haha123 
cZxid = 0x100000002
ctime = Wed Mar 21 20:01:02 CST 2018
mZxid = 0x100000004
mtime = Wed Mar 21 20:04:10 CST 2018
pZxid = 0x100000002
cversion = 0
dataVersion = 1
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 7
numChildren = 0
[zk: hadoop2:2181(CONNECTED) 7]

(6) 再次使用get命令来查看，上次修改的内容，执行命令：get /aa

[zk: hadoop2:2181(CONNECTED) 7] get /aa
haha123
cZxid = 0x100000002
ctime = Wed Mar 21 20:01:02 CST 2018
mZxid = 0x100000004
mtime = Wed Mar 21 20:04:10 CST 2018
pZxid = 0x100000002
cversion = 0
dataVersion = 1
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 7
numChildren = 0
[zk: hadoop2:2181(CONNECTED) 8]

(7) 下面我们将刚才创建的Znode删除，执行命令：delete /aa

[zk: hadoop2:2181(CONNECTED) 8] delete /aa
[zk: hadoop2:2181(CONNECTED) 9]

(8) 最后再次使用ls命令查看Zookeeper中的内容，执行命令：ls /

[zk: hadoop2:2181(CONNECTED) 9] ls /
[zookeeper]
[zk: hadoop2:2181(CONNECTED) 10]

经过验证，zk节点已经删除。

(9) 退出，执行命令：quit

[zk: hadoop2:2181(CONNECTED) 10] quit
Quitting...
2018-03-21 20:07:11,133 [myid:] - INFO  [main:ZooKeeper@684] - Session: 0x262486284b70000 closed
2018-03-21 20:07:11,139 [myid:] - INFO  [main-EventThread:ClientCnxn$EventThread@519] - EventThread shut down for session: 0x262486284b70000
[hadoop@hadoop1 ~]$

状态信息

查看一个文件的状态信息

[zk: localhost:2181(CONNECTED) 1] stat /a
cZxid = 0x200000009
ctime = Thu Mar 22 13:07:19 CST 2018
mZxid = 0x200000009
mtime = Thu Mar 22 13:07:19 CST 2018
pZxid = 0x200000009
cversion = 0
dataVersion = 0
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 1
numChildren = 0
[zk: localhost:2181(CONNECTED) 2]

详细解释：

zxid：一个事务编号，zookeeper集群内部的所有事务，都有一个全局的唯一的顺序的编号

　　它由两部分组成：就是一个 64位的长整型 long

　　高32位: 用来标识leader关系是否改变，如 0x2　　

　　低32位：用来做当前这个leader领导期间的全局的递增的事务编号，如 00000009

状态属性	说明
cZxid	数据节点创建时的事务ID
ctime	数据节点创建时的时间
mZxid	数据节点最后一次更新时的事务ID
mtime	数据节点最后一次更新时的时间
pZxid	数据节点的子节点列表最后一次被修改（是子节点列表变更，而不是子节点内容变更）时的事务ID
cversion	子节点的版本号
dataVersion	数据节点的版本号
aclVersion	数据节点的ACL版本号
ephemeralOwner	如果节点是临时节点，则表示创建该节点的会话的SessionID；如果节点是持久节点，则该属性值为0
dataLength	数据内容的长度
numChildren	数据节点当前的子节点个数

（1）修改节点a的数据，mZxid、dataVersion、dataLength 存储信息发生变化

[zk: localhost:2181(CONNECTED) 2] set /a 'aaa'
cZxid = 0x200000009
ctime = Thu Mar 22 13:07:19 CST 2018
mZxid = 0x20000000a
mtime = Thu Mar 22 13:12:53 CST 2018
pZxid = 0x200000009
cversion = 0
dataVersion = 1
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 3
numChildren = 0
[zk: localhost:2181(CONNECTED) 3]

（2）创建新的节点b，状态信息都会发生变化，zxid的事物ID也会增加

[zk: localhost:2181(CONNECTED) 5] stat /b
cZxid = 0x20000000b
ctime = Thu Mar 22 13:15:56 CST 2018
mZxid = 0x20000000b
mtime = Thu Mar 22 13:15:56 CST 2018
pZxid = 0x20000000b
cversion = 0
dataVersion = 0
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 1
numChildren = 0
[zk: localhost:2181(CONNECTED) 6]

（3）在a节点下面新增节点c，pZxid、cversion、numChildren 发生改变

[zk: localhost:2181(CONNECTED) 6] create /a/c 'c'
Created /a/c
[zk: localhost:2181(CONNECTED) 7] stat /a
cZxid = 0x200000009
ctime = Thu Mar 22 13:07:19 CST 2018
mZxid = 0x20000000a
mtime = Thu Mar 22 13:12:53 CST 2018
pZxid = 0x20000000c
cversion = 1
dataVersion = 1
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 3
numChildren = 1
[zk: localhost:2181(CONNECTED) 8]

（4）ephemeralOwner 持久性的节点信息是0x0临时的几点信息是本次会话的sessionid，如图

(5) 将leader干掉，此时第二台机器成为leader，重新创建一个文件y，此时发现czxid的前3位和之前发生变化，说明换了leader

[zk: localhost:2181(CONNECTED) 0] create /y 'yy'
Created /y
[zk: localhost:2181(CONNECTED) 2] stat /y
cZxid = 0x300000003
ctime = Thu Mar 22 13:25:51 CST 2018
mZxid = 0x300000003
mtime = Thu Mar 22 13:25:51 CST 2018
pZxid = 0x300000003
cversion = 0
dataVersion = 0
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 2
numChildren = 0
[zk: localhost:2181(CONNECTED) 3]