zookeeper 问题 - 如果领导者被杀，需要 15 分钟才能恢复
zookeeper issue - taking 15 minutes to recover if leader is killed
我正在尝试在我的网络中使用 zookeeper 实现 Kafka，但我遇到了 zookeeper 的一个奇怪问题。我环顾四周 google 并意识到许多其他用户报告了此类问题，但没有人为此发布任何适当的解决方案。
我当前的设置有 3 个不同的 zookeeper 节点（32 GB ram 专用盒子）
问题是，如果我杀死 zookeeper leader，其余的两个跟随者节点也会关闭，并且它们至少在接下来的 15-20 分钟内不会恢复。
我在 zookeeper 日志中得到的只是 "notification timeouts" 没有任何解释
这是我的动物园管理员配置文件
tickTime=2000
initLimit=10
syncLimit=5
maxClientCnxns=100
maxSessionTimeout=50000
dataDir=/var/lib/zookeeper
clientPort=2181
autopurge.snapRetainCount=100
autopurge.purgeInterval=1
preAllocSize=131072
snapCount=3000000
server.1=zo1:2888:3888
server.2=zo2:2888:3888
server.3=zo3:2888:3888
在我的 /etc/hosts 文件中，我已将 zo1 、 zo2 、 zo3 映射到它们的 ip 地址。
注意：我也通过将当前节点 ip 设置为 0.0.0.0 进行了测试，它没有任何区别。
就在几分钟前，我对其进行了测试，但再次无法恢复。
因为我有三个节点集群 zo1 、 zo2 和 zo3 。 zo3 是领导者，zo1 和 zo2 是追随者。在我杀死 zo3 节点之后。大约需要 13 分钟才能自动恢复。我在 zo1 和 zo2 中得到了以下日志。
记录 zo1。
尾/var/lib/zookeeper/zookeeper.out -n 10000 | grep 'QuorumPeer'
2019-01-02 10:25:50,848 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FollowerZooKeeperServer@140] - Shutting down
2019-01-02 10:25:50,848 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@505] - shutting down
2019-01-02 10:25:50,848 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FollowerRequestProcessor@107] - Shutting down
2019-01-02 10:25:50,848 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:CommitProcessor@184] - Shutting down
2019-01-02 10:25:50,848 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FinalRequestProcessor@402] - shutdown of request processor complete
2019-01-02 10:25:50,849 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:SyncRequestProcessor@208] - Shutting down
2019-01-02 10:25:50,849 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:QuorumPeer@865] - LOOKING
2019-01-02 10:25:50,850 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@818] - New election. My id = 1, proposed zxid=0x2d00035c8e
2019-01-02 10:25:51,057 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 400
2019-01-02 10:25:51,458 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 800
2019-01-02 10:25:52,259 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 1600
2019-01-02 10:25:53,859 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 3200
2019-01-02 10:25:57,060 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 6400
2019-01-02 10:26:03,461 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 12800
2019-01-02 10:26:16,262 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 25600
2019-01-02 10:26:41,862 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 51200
2019-01-02 10:27:33,063 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:28:33,065 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:29:33,066 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:30:33,066 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:31:33,067 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:32:33,068 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:33:33,069 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:34:33,069 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:35:33,070 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:36:33,071 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:37:33,071 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:38:33,072 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:39:33,073 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:40:33,074 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:41:33,075 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:42:33,076 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:43:33,076 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:43:33,082 [myid:1] - INFO [WorkerSender[myid=1]:QuorumPeer$QuorumServer@167] - Resolved hostname: zo3 to address: zo3/144.76.xxx.xxx
2019-01-02 10:43:33,091 [myid:1] - INFO [WorkerSender[myid=1]:QuorumPeer$QuorumServer@167] - Resolved hostname: zo3 to address: zo3/144.76.xxx.xxx
2019-01-02 10:43:33,290 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:QuorumPeer@935] - FOLLOWING
2019-01-02 10:43:33,290 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@173] - Created server with tickTime 2000 minSessionTimeout 4000 maxSessionTimeout 50000 datadir /var/lib/zookeeper/version-2 snapdir /var/lib/zookeeper/version-2
2019-01-02 10:43:33,291 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:Follower@64] - FOLLOWING - LEADER ELECTION TOOK - 1062441
2019-01-02 10:43:33,291 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:QuorumPeer$QuorumServer@167] - Resolved hostname: zo2 to address: zo2/88.198.35.34
2019-01-02 10:43:33,294 [myid:1] - WARN [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:Learner@237] - Unexpected exception, tries=0, connecting to zo2/88.198.35.34:2888
at org.apache.zookeeper.server.quorum.QuorumPeer.run(QuorumPeer.java:937)
2019-01-02 10:43:34,468 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:Learner@332] - Getting a diff from the leader 0x2d00035c8e
2019-01-02 10:43:35,120 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@687] - Established session 0x2680a49e3dc0013 with negotiated timeout 6000 for client /5.9.xxx.xxx:36664
2019-01-02 10:43:35,244 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@687] - Established session 0x1680a49b6b90011 with negotiated timeout 30000 for client /5.9.xxx.xxx:36668
2019-01-02 10:43:35,625 [myid:1] - WARN [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:Follower@118] - Got zxid 0x2e00000001 expected 0x1
来自节点 zo2 的日志，后来成为领导者
2019-01-02 10:25:50,852 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:NIOServerCnxn@1044] - Closed socket connection for client /5.9.xxx.xxx:21218 which had sessionid 0x2680a49e3dc0012
2019-01-02 10:25:50,852 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FollowerZooKeeperServer@140] - Shutting down
2019-01-02 10:25:50,853 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@505] - shutting down
2019-01-02 10:25:50,853 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FollowerRequestProcessor@107] - Shutting down
2019-01-02 10:25:50,854 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:CommitProcessor@184] - Shutting down
2019-01-02 10:25:50,854 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FinalRequestProcessor@402] - shutdown of request processor complete
2019-01-02 10:25:50,856 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:SyncRequestProcessor@208] - Shutting down
2019-01-02 10:25:50,857 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:QuorumPeer@865] - LOOKING
2019-01-02 10:25:50,858 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@818] - New election. My id = 2, proposed zxid=0x2d00035c8e
2019-01-02 10:25:51,061 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 400
2019-01-02 10:25:51,462 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 800
2019-01-02 10:25:52,283 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 1600
2019-01-02 10:25:53,884 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 3200
2019-01-02 10:25:57,084 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 6400
2019-01-02 10:26:03,485 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 12800
2019-01-02 10:26:16,286 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 25600
2019-01-02 10:26:41,887 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 51200
2019-01-02 10:27:33,087 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:28:33,088 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:29:33,089 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:30:33,090 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:31:33,091 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:32:33,092 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:33:33,092 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:34:33,093 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:35:33,094 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:36:33,095 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:37:33,095 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:38:33,096 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:39:33,097 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:40:33,098 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:41:33,099 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:42:33,100 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:43:33,293 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:QuorumPeer@947] - LEADING
2019-01-02 10:43:33,299 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:Leader@62] - TCP NoDelay set to: true
2019-01-02 10:43:33,301 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@173] - Created server with tickTime 2000 minSessionTimeout 4000 maxSessionTimeout 50000 datadir /var/lib/zookeeper/version-2 snapdir /var/lib/zookeeper/version-2
2019-01-02 10:43:33,301 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:Leader@371] - LEADING - LEADER ELECTION TOOK - 1062443
2019-01-02 10:43:34,307 [myid:2] - INFO [LearnerHandler-/144.76.120.143:64542:LearnerHandler@346] - Follower sid: 1 : info : org.apache.zookeeper.server.quorum.QuorumPeer$QuorumServer@33d2c290
2019-01-02 10:43:34,509 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:Leader@961] - Have quorum of supporters, sids: [ 1,2 ]; starting up and setting last processed zxid: 0x2e00000000
如您所见，我得到的只是日志中的连续超时，没有任何解释。
一个多星期以来一直在测试它仍然找不到任何解决方案。
如果有人能指出正确的方向，我将不胜感激。
谢谢
集群大小为 2 时不会进行选举。如果您打算 运行 HA Zookeeper 集群，请将您的 zookeeper 数量增加到 5。此外，Zookeeper 不需要 32GB 来 运行 有效。
查看：
https://docs.confluent.io/current/zookeeper/deployment.html#multi-node-setup 集群信息
和
https://docs.confluent.io/current/zookeeper/deployment.html#jvm 用于 JVM 大小调整。
致所有面临同样问题的人 - 原因是动物园管理员无所事事。如果 zookeeper 节点由于节点之间没有 activity（数据事务）而变得空闲，那么在内部 zookeeper 节点将停止相互通信。如果任何 zookeeper 节点出现故障，其他节点将不会知道，直到它们在 10-15 分钟后进行定期 ping。
我们遇到了同样的问题，发现 zookeeper leader 选举发生在 15 分钟后。
我们通过在 zookeeper 属性中设置 tcpKeepAlive=true 来绕过这 15 分钟的空闲时间。
我正在尝试在我的网络中使用 zookeeper 实现 Kafka，但我遇到了 zookeeper 的一个奇怪问题。我环顾四周 google 并意识到许多其他用户报告了此类问题，但没有人为此发布任何适当的解决方案。
我当前的设置有 3 个不同的 zookeeper 节点（32 GB ram 专用盒子） 问题是，如果我杀死 zookeeper leader，其余的两个跟随者节点也会关闭，并且它们至少在接下来的 15-20 分钟内不会恢复。
我在 zookeeper 日志中得到的只是 "notification timeouts" 没有任何解释
这是我的动物园管理员配置文件
tickTime=2000
initLimit=10
syncLimit=5
maxClientCnxns=100
maxSessionTimeout=50000
dataDir=/var/lib/zookeeper
clientPort=2181
autopurge.snapRetainCount=100
autopurge.purgeInterval=1
preAllocSize=131072
snapCount=3000000
server.1=zo1:2888:3888
server.2=zo2:2888:3888
server.3=zo3:2888:3888
在我的 /etc/hosts 文件中，我已将 zo1 、 zo2 、 zo3 映射到它们的 ip 地址。
注意：我也通过将当前节点 ip 设置为 0.0.0.0 进行了测试，它没有任何区别。
就在几分钟前，我对其进行了测试，但再次无法恢复。 因为我有三个节点集群 zo1 、 zo2 和 zo3 。 zo3 是领导者，zo1 和 zo2 是追随者。在我杀死 zo3 节点之后。大约需要 13 分钟才能自动恢复。我在 zo1 和 zo2 中得到了以下日志。
记录 zo1。
尾/var/lib/zookeeper/zookeeper.out -n 10000 | grep 'QuorumPeer'
2019-01-02 10:25:50,848 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FollowerZooKeeperServer@140] - Shutting down
2019-01-02 10:25:50,848 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@505] - shutting down
2019-01-02 10:25:50,848 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FollowerRequestProcessor@107] - Shutting down
2019-01-02 10:25:50,848 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:CommitProcessor@184] - Shutting down
2019-01-02 10:25:50,848 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FinalRequestProcessor@402] - shutdown of request processor complete
2019-01-02 10:25:50,849 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:SyncRequestProcessor@208] - Shutting down
2019-01-02 10:25:50,849 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:QuorumPeer@865] - LOOKING
2019-01-02 10:25:50,850 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@818] - New election. My id = 1, proposed zxid=0x2d00035c8e
2019-01-02 10:25:51,057 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 400
2019-01-02 10:25:51,458 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 800
2019-01-02 10:25:52,259 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 1600
2019-01-02 10:25:53,859 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 3200
2019-01-02 10:25:57,060 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 6400
2019-01-02 10:26:03,461 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 12800
2019-01-02 10:26:16,262 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 25600
2019-01-02 10:26:41,862 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 51200
2019-01-02 10:27:33,063 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:28:33,065 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:29:33,066 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:30:33,066 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:31:33,067 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:32:33,068 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:33:33,069 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:34:33,069 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:35:33,070 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:36:33,071 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:37:33,071 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:38:33,072 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:39:33,073 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:40:33,074 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:41:33,075 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:42:33,076 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:43:33,076 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:43:33,082 [myid:1] - INFO [WorkerSender[myid=1]:QuorumPeer$QuorumServer@167] - Resolved hostname: zo3 to address: zo3/144.76.xxx.xxx
2019-01-02 10:43:33,091 [myid:1] - INFO [WorkerSender[myid=1]:QuorumPeer$QuorumServer@167] - Resolved hostname: zo3 to address: zo3/144.76.xxx.xxx
2019-01-02 10:43:33,290 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:QuorumPeer@935] - FOLLOWING
2019-01-02 10:43:33,290 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@173] - Created server with tickTime 2000 minSessionTimeout 4000 maxSessionTimeout 50000 datadir /var/lib/zookeeper/version-2 snapdir /var/lib/zookeeper/version-2
2019-01-02 10:43:33,291 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:Follower@64] - FOLLOWING - LEADER ELECTION TOOK - 1062441
2019-01-02 10:43:33,291 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:QuorumPeer$QuorumServer@167] - Resolved hostname: zo2 to address: zo2/88.198.35.34
2019-01-02 10:43:33,294 [myid:1] - WARN [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:Learner@237] - Unexpected exception, tries=0, connecting to zo2/88.198.35.34:2888
at org.apache.zookeeper.server.quorum.QuorumPeer.run(QuorumPeer.java:937)
2019-01-02 10:43:34,468 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:Learner@332] - Getting a diff from the leader 0x2d00035c8e
2019-01-02 10:43:35,120 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@687] - Established session 0x2680a49e3dc0013 with negotiated timeout 6000 for client /5.9.xxx.xxx:36664
2019-01-02 10:43:35,244 [myid:1] - INFO [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@687] - Established session 0x1680a49b6b90011 with negotiated timeout 30000 for client /5.9.xxx.xxx:36668
2019-01-02 10:43:35,625 [myid:1] - WARN [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:Follower@118] - Got zxid 0x2e00000001 expected 0x1
来自节点 zo2 的日志，后来成为领导者
2019-01-02 10:25:50,852 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:NIOServerCnxn@1044] - Closed socket connection for client /5.9.xxx.xxx:21218 which had sessionid 0x2680a49e3dc0012
2019-01-02 10:25:50,852 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FollowerZooKeeperServer@140] - Shutting down
2019-01-02 10:25:50,853 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@505] - shutting down
2019-01-02 10:25:50,853 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FollowerRequestProcessor@107] - Shutting down
2019-01-02 10:25:50,854 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:CommitProcessor@184] - Shutting down
2019-01-02 10:25:50,854 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FinalRequestProcessor@402] - shutdown of request processor complete
2019-01-02 10:25:50,856 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:SyncRequestProcessor@208] - Shutting down
2019-01-02 10:25:50,857 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:QuorumPeer@865] - LOOKING
2019-01-02 10:25:50,858 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@818] - New election. My id = 2, proposed zxid=0x2d00035c8e
2019-01-02 10:25:51,061 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 400
2019-01-02 10:25:51,462 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 800
2019-01-02 10:25:52,283 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 1600
2019-01-02 10:25:53,884 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 3200
2019-01-02 10:25:57,084 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 6400
2019-01-02 10:26:03,485 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 12800
2019-01-02 10:26:16,286 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 25600
2019-01-02 10:26:41,887 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 51200
2019-01-02 10:27:33,087 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:28:33,088 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:29:33,089 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:30:33,090 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:31:33,091 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:32:33,092 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:33:33,092 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:34:33,093 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:35:33,094 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:36:33,095 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:37:33,095 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:38:33,096 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:39:33,097 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:40:33,098 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:41:33,099 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:42:33,100 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:FastLeaderElection@852] - Notification time out: 60000
2019-01-02 10:43:33,293 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:QuorumPeer@947] - LEADING
2019-01-02 10:43:33,299 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:Leader@62] - TCP NoDelay set to: true
2019-01-02 10:43:33,301 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:ZooKeeperServer@173] - Created server with tickTime 2000 minSessionTimeout 4000 maxSessionTimeout 50000 datadir /var/lib/zookeeper/version-2 snapdir /var/lib/zookeeper/version-2
2019-01-02 10:43:33,301 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:Leader@371] - LEADING - LEADER ELECTION TOOK - 1062443
2019-01-02 10:43:34,307 [myid:2] - INFO [LearnerHandler-/144.76.120.143:64542:LearnerHandler@346] - Follower sid: 1 : info : org.apache.zookeeper.server.quorum.QuorumPeer$QuorumServer@33d2c290
2019-01-02 10:43:34,509 [myid:2] - INFO [QuorumPeer[myid=2]/0:0:0:0:0:0:0:0:2181:Leader@961] - Have quorum of supporters, sids: [ 1,2 ]; starting up and setting last processed zxid: 0x2e00000000
如您所见，我得到的只是日志中的连续超时，没有任何解释。 一个多星期以来一直在测试它仍然找不到任何解决方案。
如果有人能指出正确的方向，我将不胜感激。
谢谢
集群大小为 2 时不会进行选举。如果您打算 运行 HA Zookeeper 集群，请将您的 zookeeper 数量增加到 5。此外，Zookeeper 不需要 32GB 来 运行 有效。
查看：
https://docs.confluent.io/current/zookeeper/deployment.html#multi-node-setup 集群信息
和
https://docs.confluent.io/current/zookeeper/deployment.html#jvm 用于 JVM 大小调整。
致所有面临同样问题的人 - 原因是动物园管理员无所事事。如果 zookeeper 节点由于节点之间没有 activity（数据事务）而变得空闲，那么在内部 zookeeper 节点将停止相互通信。如果任何 zookeeper 节点出现故障，其他节点将不会知道，直到它们在 10-15 分钟后进行定期 ping。
我们遇到了同样的问题，发现 zookeeper leader 选举发生在 15 分钟后。
我们通过在 zookeeper 属性中设置 tcpKeepAlive=true 来绕过这 15 分钟的空闲时间。