【发布时间】：2015-03-10 21:57:45
【问题描述】：
我在 AWS 上有一个由 Cloudera 管理的 hbase (0.96.1.1-cdh5.0.2) 集群，其中包含 4 个区域服务器和 1 个 Zookeeper 服务器。 zookeeper 服务器与 hbase master 运行在同一台主机上。我面临的问题是 3/4 区域服务器已关闭，因为它们无法连接到 zookeeper。唯一保持运行的区域服务器是与 master 和 zookeeper 在同一主机上运行的区域服务器。以下是失败区域服务器日志之一的相关部分。
2014-11-14 15:46:59,871 INFO org.apache.zookeeper.ZooKeeper: Initiating client connection, connectString=ip-10-146-188-157.ec2.internal:2181 sessionTimeout=60000 watcher=regionserver:60020, quorum=ip-10-146-188-157.ec2.internal:2181, baseZNode=/hbase
2014-11-14 15:46:59,915 INFO org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper: Process identifier=regionserver:60020 connecting to ZooKeeper ensemble=ip-10-146-188-157.ec2.internal:2181
2014-11-14 15:46:59,920 INFO org.apache.zookeeper.ClientCnxn: Opening socket connection to server ip-10-146-188-157.ec2.internal/10.146.188.157:2181. Will not attempt to authenticate using SASL (unknown error)
2014-11-14 15:47:00,649 INFO org.apache.hadoop.hbase.regionserver.ShutdownHook: Installed shutdown hook thread: Shutdownhook:regionserver60020
2014-11-14 15:47:59,948 INFO org.apache.zookeeper.ClientCnxn: Client session timed out, have not heard from server in 60041ms for sessionid 0x0, closing socket connection and attempting reconnect
2014-11-14 15:48:00,067 WARN org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper: Possibly transient ZooKeeper, quorum=ip-10-146-188-157.ec2.internal:2181, exception=org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /hbase/master
2014-11-14 15:48:00,072 INFO org.apache.hadoop.hbase.util.RetryCounter: Sleeping 1000ms before retry #0...
2014-11-14 15:48:01,067 INFO org.apache.zookeeper.ClientCnxn: Opening socket connection to server ip-10-146-188-157.ec2.internal/10.146.188.157:2181. Will not attempt to authenticate using SASL (unknown error)
2014-11-14 15:49:00,123 INFO org.apache.zookeeper.ClientCnxn: Client session timed out, have not heard from server in 60057ms for sessionid 0x0, closing socket connection and attempting reconnect
2014-11-14 15:49:00,224 WARN org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper: Possibly transient ZooKeeper, quorum=ip-10-146-188-157.ec2.internal:2181, exception=org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /hbase/master
2014-11-14 15:49:00,224 INFO org.apache.hadoop.hbase.util.RetryCounter: Sleeping 2000ms before retry #1...
2014-11-14 15:49:01,224 INFO org.apache.zookeeper.ClientCnxn: Opening socket connection to server ip-10-146-188-157.ec2.internal/10.146.188.157:2181. Will not attempt to authenticate using SASL (unknown error)
2014-11-14 15:50:00,259 INFO org.apache.zookeeper.ClientCnxn: Client session timed out, have not heard from server in 60035ms for sessionid 0x0, closing socket connection and attempting reconnect
2014-11-14 15:50:00,360 WARN org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper: Possibly transient ZooKeeper, quorum=ip-10-146-188-157.ec2.internal:2181, exception=org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /hbase/master
2014-11-14 15:50:00,360 INFO org.apache.hadoop.hbase.util.RetryCounter: Sleeping 4000ms before retry #2...
2014-11-14 15:50:01,360 INFO org.apache.zookeeper.ClientCnxn: Opening socket connection to server ip-10-146-188-157.ec2.internal/10.146.188.157:2181. Will not attempt to authenticate using SASL (unknown error)
2014-11-14 15:51:00,408 INFO org.apache.zookeeper.ClientCnxn: Client session timed out, have not heard from server in 60048ms for sessionid 0x0, closing socket connection and attempting reconnect
2014-11-14 15:51:00,509 WARN org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper: Possibly transient ZooKeeper, quorum=ip-10-146-188-157.ec2.internal:2181, exception=org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /hbase/master
2014-11-14 15:51:00,509 INFO org.apache.hadoop.hbase.util.RetryCounter: Sleeping 8000ms before retry #3...
2014-11-14 15:51:01,509 INFO org.apache.zookeeper.ClientCnxn: Opening socket connection to server ip-10-146-188-157.ec2.internal/10.146.188.157:2181. Will not attempt to authenticate using SASL (unknown error)
2014-11-14 15:52:00,559 INFO org.apache.zookeeper.ClientCnxn: Client session timed out, have not heard from server in 60051ms for sessionid 0x0, closing socket connection and attempting reconnect
2014-11-14 15:52:00,659 WARN org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper: Possibly transient ZooKeeper, quorum=ip-10-146-188-157.ec2.internal:2181, exception=org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /hbase/master
2014-11-14 15:52:00,660 ERROR org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper: ZooKeeper exists failed after 4 attempts
2014-11-14 15:52:00,661 WARN org.apache.hadoop.hbase.zookeeper.ZKUtil: regionserver:60020, quorum=ip-10-146-188-157.ec2.internal:2181, baseZNode=/hbase Unable to set watcher on znode /hbase/master
org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /hbase/master
at org.apache.zookeeper.KeeperException.create(KeeperException.java:99)
at org.apache.zookeeper.KeeperException.create(KeeperException.java:51)
at org.apache.zookeeper.ZooKeeper.exists(ZooKeeper.java:1041)
at org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper.exists(RecoverableZooKeeper.java:199)
at org.apache.hadoop.hbase.zookeeper.ZKUtil.watchAndCheckExists(ZKUtil.java:425)
at org.apache.hadoop.hbase.zookeeper.ZooKeeperNodeTracker.start(ZooKeeperNodeTracker.java:77)
at org.apache.hadoop.hbase.regionserver.HRegionServer.initializeZooKeeper(HRegionServer.java:671)
at org.apache.hadoop.hbase.regionserver.HRegionServer.preRegistrationInitialization(HRegionServer.java:644)
at org.apache.hadoop.hbase.regionserver.HRegionServer.run(HRegionServer.java:772)
at java.lang.Thread.run(Thread.java:744)
2014-11-14 15:52:00,687 ERROR org.apache.hadoop.hbase.zookeeper.ZooKeeperWatcher: regionserver:60020, quorum=ip-10-146-188-157.ec2.internal:2181, baseZNode=/hbase Received unexpected KeeperException, re-throwing exception
org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /hbase/master
at org.apache.zookeeper.KeeperException.create(KeeperException.java:99)
at org.apache.zookeeper.KeeperException.create(KeeperException.java:51)
at org.apache.zookeeper.ZooKeeper.exists(ZooKeeper.java:1041)
at org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper.exists(RecoverableZooKeeper.java:199)
at org.apache.hadoop.hbase.zookeeper.ZKUtil.watchAndCheckExists(ZKUtil.java:425)
at org.apache.hadoop.hbase.zookeeper.ZooKeeperNodeTracker.start(ZooKeeperNodeTracker.java:77)
at org.apache.hadoop.hbase.regionserver.HRegionServer.initializeZooKeeper(HRegionServer.java:671)
at org.apache.hadoop.hbase.regionserver.HRegionServer.preRegistrationInitialization(HRegionServer.java:644)
at org.apache.hadoop.hbase.regionserver.HRegionServer.run(HRegionServer.java:772)
at java.lang.Thread.run(Thread.java:744)
2014-11-14 15:52:00,692 FATAL org.apache.hadoop.hbase.regionserver.HRegionServer: ABORTING region server 0.0.0.0,60020,1415998019646: Unexpected exception during initialization, aborting
org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /hbase/master
at org.apache.zookeeper.KeeperException.create(KeeperException.java:99)
at org.apache.zookeeper.KeeperException.create(KeeperException.java:51)
at org.apache.zookeeper.ZooKeeper.exists(ZooKeeper.java:1041)
at org.apache.hadoop.hbase.zookeeper.RecoverableZooKeeper.exists(RecoverableZooKeeper.java:199)
at org.apache.hadoop.hbase.zookeeper.ZKUtil.watchAndCheckExists(ZKUtil.java:425)
at org.apache.hadoop.hbase.zookeeper.ZooKeeperNodeTracker.start(ZooKeeperNodeTracker.java:77)
at org.apache.hadoop.hbase.regionserver.HRegionServer.initializeZooKeeper(HRegionServer.java:671)
at org.apache.hadoop.hbase.regionserver.HRegionServer.preRegistrationInitialization(HRegionServer.java:644)
at org.apache.hadoop.hbase.regionserver.HRegionServer.run(HRegionServer.java:772)
at java.lang.Thread.run(Thread.java:744)
我怀疑这可能与 /etc/hosts 配置有关，但无法弄清楚问题所在。集群中每个实例的 /etc/hosts 是：
127.0.0.1 localhost.localdomain localhost
::1 localhost6.localdomain6 localhost6
处理zookeeper的hbase-site.xml部分是。
<property>
<name>zookeeper.znode.parent</name>
<value>/hbase</value>
</property>
<property>
<name>zookeeper.znode.rootserver</name>
<value>root-region-server</value>
</property>
<property>
<name>hbase.zookeeper.quorum</name>
<value>ip-10-146-188-157.ec2.internal</value>
</property>
<property>
<name>hbase.zookeeper.property.clientPort</name>
<value>2181</value>
</property>
任何帮助将不胜感激。
【问题讨论】：