一、什么是ZooKeeper?

ZooKeeper(动物园管理员),顾名思义,是用来管理Hadoop(大象)、Hive(蜜蜂)、Pig(小猪)的管理员,同时Apache HBase、Apache Solr、LinkedIn Sensei等众多项目中都采用了ZooKeeper。

ZooKeeper是一个集中式服务,用于维护配置信息、命名、提供分布式同步和提供组服务。所有这些类型的服务都以某种形式被分布式应用程序使用。每次它们被实现时,都有大量的工作用于修复不可避免的bug和竞选条件。由于实现这类服务的困难,应用程序最初通常跳过它们,这使得它们在存在更改时变得脆弱,并且难以管理。即使正确完成,这些服务的不同实现也会在部署应用程序时导致管理复杂性。

二、Zookeeper集群

Kafka使用Zookeeper来存储集群元数据以及消费者元数据。Zookeeper集群被称为群组。Zookeeper使用的是一致性协议,所以建议每个群组里应该包含奇数个节点(比如3个、5个等),因为只有当群组里的大多数节点(也就是法定人数)处于可用状态,Zookeeper才能处理外部的请求。也就是说,如果3个节点的集群,允许1个节点失效。如果5个节点的集群,允许2个节点失效。

三、Docker方式配置ZooKeeper

myid文件:标识要写到快照目录下面myid文件里

采用Docker镜像:wurstmeister/zookeeper,部署一个3节点的集群。

节点一:

  • docker-compose.yml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
version: '2'
services:
zookeeper:
image: wurstmeister/zookeeper
volumes:
- ./zoo.cfg:/opt/zookeeper-3.4.9/conf/zoo.cfg
- ./data:/opt/zookeeper-3.4.9/data
- ./logs:/opt/zookeeper-3.4.9/logs
- ./myid:/opt/zookeeper-3.4.9/data/myid
ports:
- "2181:2181"
- "2888:2888"
- "3888:3888"
restart: always
  • myid文件

    1
    1
  • zoo.cfg文件

    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
    # The number of milliseconds of each tick
    tickTime=2000
    # The number of ticks that the initial
    # synchronization phase can take
    initLimit=10
    # The number of ticks that can pass between
    # sending a request and getting an acknowledgement
    syncLimit=5
    # the directory where the snapshot is stored.
    # do not use /tmp for storage, /tmp here is just
    # example sakes.
    dataDir=/opt/zookeeper-3.4.9/data
    dataLogDir=/opt/zookeeper-3.4.9/logs
    # the port at which the clients will connect
    clientPort=2181
    # the maximum number of client connections.
    # increase this if you need to handle more clients
    #maxClientCnxns=60
    #
    # Be sure to read the maintenance section of the
    # administrator guide before turning on autopurge.
    #
    # http://zookeeper.apache.org/doc/current/zookeeperAdmin.html#sc_maintenance
    #
    # The number of snapshots to retain in dataDir
    autopurge.snapRetainCount=3
    # Purge task interval in hours
    # Set to "0" to disable auto purge feature
    autopurge.purgeInterval=1
    #kafka01 (10.100.158.245)
    server.1=0.0.0.0:2888:3888
    #kafka02 (10.100.19.15)
    server.2=10.150.19.15:2888:3888
    #kafka03 (10.100.124.217)
    server.3=10.100.124.217:2888:3888

节点二:

  • docker-compose.yml

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    version: '2'
    services:
    zookeeper:
    image: wurstmeister/zookeeper
    volumes:
    - ./zoo.cfg:/opt/zookeeper-3.4.9/conf/zoo.cfg
    - ./data:/opt/zookeeper-3.4.9/data
    - ./logs:/opt/zookeeper-3.4.9/logs
    - ./myid:/opt/zookeeper-3.4.9/data/myid
    ports:
    - "2180:2181"
    - "2888:2888"
    - "3888:3888"
    restart: always
  • myid文件

    1
    2
  • zoo.cfg文件

    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
    # The number of milliseconds of each tick
    tickTime=2000
    # The number of ticks that the initial
    # synchronization phase can take
    initLimit=10
    # The number of ticks that can pass between
    # sending a request and getting an acknowledgement
    syncLimit=5
    # the directory where the snapshot is stored.
    # do not use /tmp for storage, /tmp here is just
    # example sakes.
    dataDir=/opt/zookeeper-3.4.9/data
    dataLogDir=/opt/zookeeper-3.4.9/logs
    # the port at which the clients will connect
    clientPort=2181
    # the maximum number of client connections.
    # increase this if you need to handle more clients
    #maxClientCnxns=60
    #
    # Be sure to read the maintenance section of the
    # administrator guide before turning on autopurge.
    #
    # http://zookeeper.apache.org/doc/current/zookeeperAdmin.html#sc_maintenance
    #
    # The number of snapshots to retain in dataDir
    autopurge.snapRetainCount=3
    # Purge task interval in hours
    # Set to "0" to disable auto purge feature
    autopurge.purgeInterval=1
    #kafka01 (10.100.158.245)
    server.1=10.100.158.245:2888:3888
    #kafka02 (10.100.19.15)
    server.2=0.0.0.0:2888:3888
    #kafka03 (10.100.124.217)
    server.3=10.100.124.217:2888:3888

节点三:

  • docker-compose.yml

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    version: '2'
    services:
    zookeeper:
    image: wurstmeister/zookeeper
    volumes:
    - ./zoo.cfg:/opt/zookeeper-3.4.9/conf/zoo.cfg
    - ./data:/opt/zookeeper-3.4.9/data
    - ./logs:/opt/zookeeper-3.4.9/logs
    - ./myid:/opt/zookeeper-3.4.9/data/myid
    ports:
    - "2179:2181"
    - "2888:2888"
    - "3888:3888"
    restart: always
  • myid文件

    1
    3
  • zoo.cfg文件

    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
    # The number of milliseconds of each tick
    tickTime=2000
    # The number of ticks that the initial
    # synchronization phase can take
    initLimit=10
    # The number of ticks that can pass between
    # sending a request and getting an acknowledgement
    syncLimit=5
    # the directory where the snapshot is stored.
    # do not use /tmp for storage, /tmp here is just
    # example sakes.
    dataDir=/opt/zookeeper-3.4.9/data
    dataLogDir=/opt/zookeeper-3.4.9/logs
    # the port at which the clients will connect
    clientPort=2181
    # the maximum number of client connections.
    # increase this if you need to handle more clients
    #maxClientCnxns=60
    #
    # Be sure to read the maintenance section of the
    # administrator guide before turning on autopurge.
    #
    # http://zookeeper.apache.org/doc/current/zookeeperAdmin.html#sc_maintenance
    #
    # The number of snapshots to retain in dataDir
    autopurge.snapRetainCount=3
    # Purge task interval in hours
    # Set to "0" to disable auto purge feature
    autopurge.purgeInterval=1
    #kafka01 (10.100.158.245)
    server.1=10.100.158.245:2888:3888
    #kafka02 (10.100.19.15)
    server.2=10.100.19.15:2888:3888
    #kafka03 (10.100.124.217)
    server.3=0.0.0.0:2888:3888

四、查看ZooKeeper状态

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
# 服务器1
[root@VM_158_24_centos ~]# docker exec -it s1_zookeeper_1 bin/zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /opt/zookeeper-3.4.9/bin/../conf/zoo.cfg
Mode: leader
#服务器2
[root@VM_19_15_centos ~]# docker exec -it s2_zookeeper_1 bin/zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /opt/zookeeper-3.4.9/bin/../conf/zoo.cfg
Mode: follower
#服务器3
[root@VM_124_27_centos ~]# docker exec -it s3_zookeeper_1 bin/zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /opt/zookeeper-3.4.9/bin/../conf/zoo.cfg
Mode: follower

停掉leader,2个follower的状态有变化,如下:

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
58
59
60
61
62
63
64
65
66
67
68
69
70
71
# 服务器1
[root@VM_158_245_centos ~]# docker stop s1_zookeeper_1
s1_zookeeper_1
# 服务器2
2018-10-30 10:54:44,555 [myid:2] - WARN [WorkerSender[myid=2]:QuorumCnxManager@400] - Cannot open channel to 1 at election address /10.154.158.245:3888
java.net.ConnectException: Connection refused
at java.net.PlainSocketImpl.socketConnect(Native Method)
at java.net.AbstractPlainSocketImpl.doConnect(AbstractPlainSocketImpl.java:339)
at java.net.AbstractPlainSocketImpl.connectToAddress(AbstractPlainSocketImpl.java:200)
at java.net.AbstractPlainSocketImpl.connect(AbstractPlainSocketImpl.java:182)
at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:392)
at java.net.Socket.connect(Socket.java:579)
at org.apache.zookeeper.server.quorum.QuorumCnxManager.connectOne(QuorumCnxManager.java:381)
at org.apache.zookeeper.server.quorum.QuorumCnxManager.toSend(QuorumCnxManager.java:354)
at org.apache.zookeeper.server.quorum.FastLeaderElection$Messenger$WorkerSender.process(FastLeaderElection.java:452)
at org.apache.zookeeper.server.quorum.FastLeaderElection$Messenger$WorkerSender.run(FastLeaderElection.java:433)
at java.lang.Thread.run(Thread.java:745)
2018-10-30 10:54:44,556 [myid:2] - INFO [WorkerSender[myid=2]:QuorumPeer$QuorumServer@149] - Resolved hostname: 10.154.158.245 to address: /10.154.158.245
2018-10-30 10:54:44,556 [myid:2] - INFO [WorkerReceiver[myid=2]:FastLeaderElection@600] - Notification: 1 (message format version), 3 (n.leader), 0x500000000 (n.zxid), 0xf4 (n.round), LOOKING (n.state), 2 (n.sid), 0x5 (n.peerEpoch) LOOKING (my state)
2018-10-30 10:54:44,759 [myid:2] - INFO [QuorumPeer[myid=2]/0.0.0.0:2181:QuorumPeer@844] - FOLLOWING
2018-10-30 10:54:44,759 [myid:2] - INFO [QuorumPeer[myid=2]/0.0.0.0:2181:ZooKeeperServer@173] - Created server with tickTime 2000 minSessionTimeout 4000 maxSessionTimeout 40000 datadir /opt/zookeeper-3.4.9/logs/version-2 snapdir /opt/zookeeper-3.4.9/data/version-2
2018-10-30 10:54:44,759 [myid:2] - INFO [QuorumPeer[myid=2]/0.0.0.0:2181:Follower@61] - FOLLOWING - LEADER ELECTION TOOK - 215
2018-10-30 10:54:44,760 [myid:2] - INFO [QuorumPeer[myid=2]/0.0.0.0:2181:QuorumPeer$QuorumServer@149] - Resolved hostname: 10.105.124.217 to address: /10.105.124.217
2018-10-30 10:54:44,761 [myid:2] - WARN [QuorumPeer[myid=2]/0.0.0.0:2181:Learner@236] - Unexpected exception, tries=0, connecting to /10.105.124.217:2888
java.net.ConnectException: Connection refused
at java.net.PlainSocketImpl.socketConnect(Native Method)
at java.net.AbstractPlainSocketImpl.doConnect(AbstractPlainSocketImpl.java:339)
at java.net.AbstractPlainSocketImpl.connectToAddress(AbstractPlainSocketImpl.java:200)
at java.net.AbstractPlainSocketImpl.connect(AbstractPlainSocketImpl.java:182)
at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:392)
at java.net.Socket.connect(Socket.java:579)
at org.apache.zookeeper.server.quorum.Learner.connectToLeader(Learner.java:228)
at org.apache.zookeeper.server.quorum.Follower.followLeader(Follower.java:69)
at org.apache.zookeeper.server.quorum.QuorumPeer.run(QuorumPeer.java:846)
2018-10-30 10:54:45,794 [myid:2] - INFO [QuorumPeer[myid=2]/0.0.0.0:2181:Learner@329] - Getting a snapshot from leader
2018-10-30 10:54:45,796 [myid:2] - INFO [QuorumPeer[myid=2]/0.0.0.0:2181:FileTxnSnapLog@240] - Snapshotting: 0x500000000 to /opt/zookeeper-3.4.9/data/version-2/snapshot.500000000
# 服务器3
2018-10-30 10:54:44,567 [myid:3] - INFO [QuorumPeer[myid=3]/0.0.0.0:2181:FastLeaderElection@818] - New election. My id = 3, proposed zxid=0x500000000
2018-10-30 10:54:44,568 [myid:3] - WARN [WorkerSender[myid=3]:QuorumCnxManager@400] - Cannot open channel to 1 at election address /10.154.158.245:3888
java.net.ConnectException: Connection refused
at java.net.PlainSocketImpl.socketConnect(Native Method)
at java.net.AbstractPlainSocketImpl.doConnect(AbstractPlainSocketImpl.java:339)
at java.net.AbstractPlainSocketImpl.connectToAddress(AbstractPlainSocketImpl.java:200)
at java.net.AbstractPlainSocketImpl.connect(AbstractPlainSocketImpl.java:182)
at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:392)
at java.net.Socket.connect(Socket.java:579)
at org.apache.zookeeper.server.quorum.QuorumCnxManager.connectOne(QuorumCnxManager.java:381)
at org.apache.zookeeper.server.quorum.QuorumCnxManager.toSend(QuorumCnxManager.java:354)
at org.apache.zookeeper.server.quorum.FastLeaderElection$Messenger$WorkerSender.process(FastLeaderElection.java:452)
at org.apache.zookeeper.server.quorum.FastLeaderElection$Messenger$WorkerSender.run(FastLeaderElection.java:433)
at java.lang.Thread.run(Thread.java:745)
2018-10-30 10:54:44,569 [myid:3] - INFO [WorkerSender[myid=3]:QuorumPeer$QuorumServer@149] - Resolved hostname: 10.154.158.245 to address: /10.154.158.245
2018-10-30 10:54:44,569 [myid:3] - INFO [WorkerReceiver[myid=3]:FastLeaderElection@600] - Notification: 1 (message format version), 3 (n.leader), 0x500000000 (n.zxid), 0xf4 (n.round), LOOKING (n.state), 3 (n.sid), 0x5 (n.peerEpoch) LOOKING (my state)
2018-10-30 10:54:44,575 [myid:3] - INFO [WorkerReceiver[myid=3]:FastLeaderElection@600] - Notification: 1 (message format version), 3 (n.leader), 0x500000000 (n.zxid), 0xf4 (n.round), LOOKING (n.state), 2 (n.sid), 0x5 (n.peerEpoch) LOOKING (my state)
2018-10-30 10:54:44,776 [myid:3] - INFO [QuorumPeer[myid=3]/0.0.0.0:2181:QuorumPeer@856] - LEADING
2018-10-30 10:54:44,779 [myid:3] - INFO [QuorumPeer[myid=3]/0.0.0.0:2181:Leader@59] - TCP NoDelay set to: true
2018-10-30 10:54:44,780 [myid:3] - INFO [QuorumPeer[myid=3]/0.0.0.0:2181:ZooKeeperServer@173] - Created server with tickTime 2000 minSessionTimeout 4000 maxSessionTimeout 40000 datadir /opt/zookeeper-3.4.9/logs/version-2 snapdir /opt/zookeeper-3.4.9/data/version-2
2018-10-30 10:54:44,780 [myid:3] - INFO [QuorumPeer[myid=3]/0.0.0.0:2181:Leader@361] - LEADING - LEADER ELECTION TOOK - 219
2018-10-30 10:54:45,782 [myid:3] - INFO [LearnerHandler-/10.154.19.15:42938:LearnerHandler@329] - Follower sid: 2 : info : org.apache.zookeeper.server.quorum.QuorumPeer$QuorumServer@7f54169
2018-10-30 10:54:45,808 [myid:3] - INFO [LearnerHandler-/10.154.19.15:42938:LearnerHandler@384] - Synchronizing with Follower sid: 2 maxCommittedLog=0x0 minCommittedLog=0x0 peerLastZxid=0x300000000
2018-10-30 10:54:45,809 [myid:3] - INFO [LearnerHandler-/10.154.19.15:42938:LearnerHandler@458] - Sending SNAP
2018-10-30 10:54:45,809 [myid:3] - INFO [LearnerHandler-/10.154.19.15:42938:LearnerHandler@482] - Sending snapshot last zxid of peer is 0x300000000 zxid of leader is 0x600000000sent zxid of db as 0x500000000
2018-10-30 10:54:45,820 [myid:3] - INFO [LearnerHandler-/10.154.19.15:42938:LearnerHandler@518] - Received NEWLEADER-ACK message from 2
2018-10-30 10:54:45,820 [myid:3] - INFO [QuorumPeer[myid=3]/0.0.0.0:2181:Leader@952] - Have quorum of supporters, sids: [ 2,3 ]; starting up and setting last processed zxid: 0x600000000

再次查看status:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
#服务器2
[root@VM_19_15_centos ~]# docker exec -it s2_zookeeper_1 bin/zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /opt/zookeeper-3.4.9/bin/../conf/zoo.cfg
Mode: follower
#服务器3
[root@VM_124_217_centos ~]# docker exec -it s3_zookeeper_1 bin/zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /opt/zookeeper-3.4.9/bin/../conf/zoo.cfg
Mode: leader

服务器3变成leader了!

重新启动服务器1,服务器1是follower

1
2
3
4
5
6
7
8
9
10
11
12
[root@VM_158_245_centos ~]# docker start s1_zookeeper_1
s1_zookeeper_1
[root@VM_158_245_centos ~]# docker exec -it s1_zookeeper_1 bin/zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /opt/zookeeper-3.4.9/bin/../conf/zoo.cfg
Mode: follower

参考

Configuring Zookeeper Cluster

多台虚拟机搭建zookeeper集群

搭建kafka集群详细教程

zookeeper.apache.org