关闭Kerberos后hbase集群RegionServer经常意外退出

按通常网上的关闭Kerberos集群的方法关闭集群后，hbase集群经常有单节点意外退出，查看日志会发现很长的报错信息，下面从头按顺序开始列举：

ABORTING region server cdh136,60020,1539306655523: org.apache.hadoop.hbase.YouAreDeadException: Server REPORT rejected; currently processing cdh136,60020,1539306655523 as dead server

RegionServer abort: loaded coprocessors are: [org.apache.hadoop.hbase.security.access.SecureBulkLoadEndpoint]

Problem shutting down provider 'org.apache.hadoop.hbase.wal.DefaultWALProvider@14ac4cd2': BP-577498931-192.168.xxx.xx-1530099044214:blk_1073933265_192620 does not exist or is not under Constructionblk_1073933265_192620{blockUCState=UNDER_RECOVERY, primaryNodeIndex=2, replicas=[ReplicaUnderConstruction[[DISK]DS-6dd36561-ef5b-40b4-8cde-d8e2aaedb29a:NORMAL:...

Shutdown / close of WAL failed: java.io.IOException: BP-577498931-192.168.2.148-1530099044214:blk_1073933265_192620 does not exist or is not under Constructionblk_1073933265_192620{blockUCState=UNDER_RECOVERY, primaryNodeIndex=2, replicas=...

ABORTING region server cdh136,60020,1539306655523: regionserver:60020-0x46664a239c20206, quorum=cdh148:2181,cdh136:2181,cdh138:2181,cdh139:2181,cdh137:2181, baseZNode=/hbase regionserver:60020-0x46664a239c20206 received expired from ZooKeeper, aborting

regionserver:60020-0x46664a239c20206, quorum=cdh148:2181,cdh136:2181,cdh138:2181,cdh139:2181,cdh137:2181, baseZNode=/hbase Received unexpected KeeperException, re-throwing exception
org.apache.zookeeper.KeeperException$SessionExpiredException: KeeperErrorCode = Session expired for /hbase/replication/rs/cdh136,60020,1539306655523

Region server exiting
java.lang.RuntimeException: HRegionServer Aborted
	at org.apache.hadoop.hbase.regionserver.HRegionServerCommandLine.start(HRegionServerCommandLine.java:68)

regionserver:60020-0x46664a239c20206, quorum=cdh148:2181,cdh136:2181,cdh138:2181,cdh139:2181,cdh137:2181, baseZNode=/hbase Received unexpected KeeperException, re-throwing exception
org.apache.zookeeper.KeeperException$SessionExpiredException: KeeperErrorCode = Session expired for /hbase/replication/rs/cdh136,60020,1539306655523

解决方法：信息涉及安全和认证的方面，很容易猜到是kerberos关闭不彻底的问题，故障最初显示由security block load异常开始，在CDH集群hbase的配置信息中搜索kerberos发现该选项：

HBase REST 身份验证
hbase.rest.authentication.type
HBase（服务范围）

• simple
• kerberos

还处于勾选kerberos的状态！改为勾选simple，稍等片刻后重新部署客户端配置并重启集群，问题解决！