ZooKeeper相关错误的解决办法记录

一、错误1

1.1、错误描述

ZooKeeper Server（“FOLLOWER和LEADER”都有）的日志中显示有以下所示错误：

[mw_shl_code=bash,true]2016-05-14 15:33:01,818 [myid:2] - ERROR [CommitProcessor:2:NIOServerCnxn@178] -
Unexpected Exception:
java.nio.channels.CancelledKeyException
        at sun.nio.ch.SelectionKeyImpl.ensureValid(SelectionKeyImpl.java:55)
        at sun.nio.ch.SelectionKeyImpl.interestOps(SelectionKeyImpl.java:59)
        at org.apache.zookeeper.server.NIOServerCnxn.sendBuffer(NIOServerCnxn.ja
va:151)
        at org.apache.zookeeper.server.NIOServerCnxn.sendResponse(NIOServerCnxn.
java:1081)
        at org.apache.zookeeper.server.FinalRequestProcessor.processRequest(Fina
lRequestProcessor.java:170)
        at org.apache.zookeeper.server.quorum.CommitProcessor.run(CommitProcessor.java:74)[/mw_shl_code]

1.2、错误原因分析

ZooKeeper Server发送回复时，Socket连接已经被关闭。

1.3、错误解决

当ZooKeeper Server发送回复时，增加一个“sk.isValid()”的判断。以上其实是一个bug，在ZooKeeper 3.4.8版本中得到修复。

1.4、其他

这个错误在上线“使用ZooKeeper获取MQ地址方案”之前也存在。

二、错误2

2.1、错误描述

ZooKeeper Server（“FOLLOWER”）日志中显示有以下所示错误，出现该错误后，作为“FOLLOWER”的该ZooKeeper Server在一段时间内会停止工作：


[mw_shl_code=bash,true]2016-05-15 04:04:40,569 [myid:1] - WARN  [SyncThread:1:FileTxnLog@334] - fsync-ing the write ahead log in SyncThread:1 took 2243ms which will adversely effect operation latency. See the
ZooKeeper troubleshooting guide[/mw_shl_code]

[mw_shl_code=bash,true]2016-05-14 15:32:50,764 [myid:1] - WARN  [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:Follower@89] - Exception when following the leader
java.io.EOFException
        at java.io.DataInputStream.readInt(DataInputStream.java:375)
        at org.apache.jute.BinaryInputArchive.readInt(BinaryInputArchive.java:63)
        at org.apache.zookeeper.server.quorum.QuorumPacket.deserialize(QuorumPacket.java:83)
        at org.apache.jute.BinaryInputArchive.readRecord(BinaryInputArchive.java:103)
        at org.apache.zookeeper.server.quorum.Learner.readPacket(Learner.java:153)
        at org.apache.zookeeper.server.quorum.Follower.followLeader(Follower.java:85)
        at org.apache.zookeeper.server.quorum.QuorumPeer.run(QuorumPeer.java:786)
2016-05-14 15:32:50,764 [myid:1] - INFO  [QuorumPeer[myid=1]/0:0:0:0:0:0:0:0:2181:Follower@166] - shutdown called
java.lang.Exception: shutdown Follower
        at org.apache.zookeeper.server.quorum.Follower.shutdown(Follower.java:166)
        at org.apache.zookeeper.server.quorum.QuorumPeer.run(QuorumPeer.java:790)[/mw_shl_code]

相应的ZooKeeper Server（“LEADER”）日志中显示有如下所示错误：
[mw_shl_code=bash,true]2016-05-14 15:32:42,605 [myid:3] - WARN  [SyncThread:3:FileTxnLog@334] - fsync-i
ng the write ahead log in SyncThread:3 took 3041ms which will adversely effect o
peration latency. See the ZooKeeper troubleshooting guide[/mw_shl_code]


[mw_shl_code=bash,true]2016-05-14 15:32:50,764 [myid:3] - WARN  [QuorumPeer[myid=3]/0:0:0:0:0:0:0:0:218
1:LearnerHandler@687] - Closing connection to peer due to transaction timeout.
2016-05-14 15:32:50,764 [myid:3] - WARN  [LearnerHandler-/10.110.20.23:39390:Lea
rnerHandler@646] - ******* GOODBYE /10.110.20.23:39390 ********
2016-05-14 15:32:50,764 [myid:3] - WARN  [LearnerHandler-/10.110.20.23:39390:Lea
rnerHandler@658] - Ignoring unexpected exception
java.lang.InterruptedException
        at java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireInterrup
tibly(AbstractQueuedSynchronizer.java:1199)
        at java.util.concurrent.locks.ReentrantLock.lockInterruptibly(ReentrantL
ock.java:312)
        at java.util.concurrent.LinkedBlockingQueue.put(LinkedBlockingQueue.java
:294)
        at org.apache.zookeeper.server.quorum.LearnerHandler.shutdown(LearnerHan
dler.java:656)
        at org.apache.zookeeper.server.quorum.LearnerHandler.run(LearnerHandler.
java:649)[/mw_shl_code]

2.2、错误原因分析

“FOLLOWER”在跟“LEADER”同步时，fsync操作时间过长，导致超时。

2.3、错误解决

增加“tickTime”或者“initLimit和syncLimit”的值，或者两者都增大。

2.4、其他

这个错误在上线“使用ZooKeeper获取MQ地址方案”之前也存在，只不过没有这么高频率，而上线了“使用ZooKeeper获取MQ地址方案”之后，ZooKeeper Server之间的同步数据量增大，ZooKeeper Server的负载加重，因而最终导致高频率出现上述错误。