自己的虚拟机集群，一次强制关机后，发现slave2的zookeeper起不来了
下午Reading configuration from: /var/run/cloudera-scm-agent/process/517-zookeeper-server/zoo.cfg下午Defaulting to majority quorums下午autopurge.snapRetainCount set to 5下午autopurge.purgeInterval set to 24下午Purge task started.下午Error occured while purging.org.apache.zookeeper.server.persistence.FileTxnSnapLog$DatadirException: Missing data directory /var/lib/zookeeper/version-2, automatic data directory creation is disabled (zookeeper.datadir.autocreate is false). Please create this directory manually.at org.<init>(FileTxnSnapLog.java:102)at org(PurgeTxnLog.java:68)at org$PurgeTask.run(DatadirCleanupManager.java:140)at java(Timer.java:512)at java(Timer.java:462)
来源： :7180/cmf/process/all/logs/context?timestamp=1437989392521&port=9000&host=slave2.cloudera&path=%2Fvar%2Flog%2Fzookeeper%2Fzookeeper-cmf-zookeeper1-SERVER-slave2.cloudera.log&roleId=113
Removing data from /var/zookeeper/version-2 then restart seems to “fix” the problem (it gets a snapshot from one of the other nodes in the quorum). This is Zookeeper 3.3.5+19.5-1~squeeze-cdh3, i.e. from Cloudera’s distribution.
来源： https://issues.apache.org/jira/browse/ZOOKEEPER-1546
看了老外的文章，，下面上是处理方法
more /etc/zookeeper/conf.dist/zoo.cfg 找到datadir [root@slave2 zookeeper]# pwd /var/lib/zookeeper [root@slave2 zookeeper]# ls myid version-2 version-2.bak 清空version-2目录下的所有文件
告诉自己，我这次失败了，重新开始吧!下次我会吸取教训，不让自己犯同样的错误的
