这两天，在新的服务器上部署大数据相关内容，其中安装HBase，所有安装都已成功，但是启动的时候，HMaster启动失败，有时候，jps查看是HMaster存在几十秒就挂了（总共七台服务器，node1为master，node2为backup-master）。查看日志，报错内容如下：
2021-08-04 15:32:38,839 INFO [main] util.FSTableDescriptors: ta', {TABLE_ATTRIBUTES => {IS_META => 'true', REGION_REPLICATION => '1', coprocessor$1 => '|org.apache.hadoop.hbase.coprocessor.MultiRowMutationEndpoint|536870911|'}}, {NAME => 'info', BLOOMFILTER => 'NONE', IN_MEMORY => 'true', VERSIONS => '3', KEEP_DELETED_CELLS => 'FALSE', DATA_BLOCK_ENCODING => 'NONE', COMPRESSION => 'NONE', TTL => 'FOREVER', MIN_VERSIONS => '0', BLOCKCACHE => 'true', BLOCKSIZE => '8192', REPLICATION_SCOPE => '0'}, {NAME => 'rep_barrier', BLOOMFILTER => 'NONE', IN_MEMORY => 'true', VERSIONS => '2147483647', KEEP_DELETED_CELLS => 'FALSE', DATA_BLOCK_ENCODING => 'NONE', COMPRESSION => 'NONE', TTL => 'FOREVER', MIN_VERSIONS => '0', BLOCKCACHE => 'true', BLOCKSIZE => '65536', REPLICATION_SCOPE => '0'}, {NAME => 'table', BLOOMFILTER => 'NONE', IN_MEMORY => 'true', VERSIONS => '3', KEEP_DELETED_CELLS => 'FALSE', DATA_BLOCK_ENCODING => 'NONE', COMPRESSION => 'NONE', TTL => 'FOREVER', MIN_VERSIONS => '0', BLOCKCACHE => 'true', BLOCKSIZE => '8192', REPLICATION_SCOPE => '0'}
2021-08-04 15:32:39,026 WARN [main] util.FSTableDescriptors: Failed cleanup of hdfs://node1:9000/user/hbase/data/hbase/meta/.tmp/.tableinfo.0000000001
2021-08-04 15:32:39,042 WARN [main] util.FSTableDescriptors: Failed cleanup of hdfs://node1:9000/user/hbase/data/hbase/meta/.tmp/.tableinfo.0000000002
2021-08-04 15:32:39,052 WARN [main] util.FSTableDescriptors: Failed cleanup of hdfs://node1:9000/user/hbase/data/hbase/meta/.tmp/.tableinfo.0000000003
2021-08-04 15:32:39,062 WARN [main] util.FSTableDescriptors: Failed cleanup of hdfs://node1:9000/user/hbase/data/hbase/meta/.tmp/.tableinfo.0000000004
2021-08-04 15:32:39,072 WARN [main] util.FSTableDescriptors: Failed cleanup of hdfs://node1:9000/user/hbase/data/hbase/meta/.tmp/.tableinfo.0000000005
2021-08-04 15:32:39,082 WARN [main] util.FSTableDescriptors: Failed cleanup of hdfs://node1:9000/user/hbase/data/hbase/meta/.tmp/.tableinfo.0000000006
2021-08-04 15:32:39,094 WARN [main] util.FSTableDescriptors: Failed cleanup of hdfs://node1:9000/user/hbase/data/hbase/meta/.tmp/.tableinfo.0000000007
2021-08-04 15:32:39,104 WARN [main] util.FSTableDescriptors: Failed cleanup of hdfs://node1:9000/user/hbase/data/hbase/meta/.tmp/.tableinfo.0000000008
2021-08-04 15:32:39,115 WARN [main] util.FSTableDescriptors: Failed cleanup of hdfs://node1:9000/user/hbase/data/hbase/meta/.tmp/.tableinfo.0000000009
2021-08-04 15:32:39,123 WARN [main] util.FSTableDescriptors: Failed cleanup of hdfs://node1:9000/user/hbase/data/hbase/meta/.tmp/.tableinfo.0000000010
2021-08-04 15:32:39,124 ERROR [main] regionserver.HRegionServer: Failed construction RegionServer
java.io.IOException: Failed update hbase:meta table descriptor
at org.apache.hadoop.hbase.util.FSTableDescriptors.tryUpdateMetaTableDescriptor(FSTableDescriptors.java:144)
at org.apache.hadoop.hbase.regionserver.HRegionServer.initializeFileSystem(HRegionServer.java:738)
at org.apache.hadoop.hbase.regionserver.HRegionServer.<init>(HRegionServer.java:635)
at org.apache.hadoop.hbase.master.HMaster.<init>(HMaster.java:528)
at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62)
at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
at java.lang.reflect.Constructor.newInstance(Constructor.java:423)
at org.apache.hadoop.hbase.master.HMaster.constructMaster(HMaster.java:3163)
at org.apache.hadoop.hbase.master.HMasterCommandLine.startMaster(HMasterCommandLine.java:253)
at org.apache.hadoop.hbase.master.HMasterCommandLine.run(HMasterCommandLine.java:149)
at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:76)
at org.apache.hadoop.hbase.util.ServerCommandLine.doMain(ServerCommandLine.java:149)
at org.apache.hadoop.hbase.master.HMaster.main(HMaster.java:3181)
2021-08-04 15:32:39,135 ERROR [main] master.HMasterCommandLine: Master exiting
java.lang.RuntimeException: Failed construction of Master: class org.apache.hadoop.hbase.master.HMaster.
at org.apache.hadoop.hbase.master.HMaster.constructMaster(HMaster.java:3170)
at org.apache.hadoop.hbase.master.HMasterCommandLine.startMaster(HMasterCommandLine.java:253)
at org.apache.hadoop.hbase.master.HMasterCommandLine.run(HMasterCommandLine.java:149)
at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:76)
at org.apache.hadoop.hbase.util.ServerCommandLine.doMain(ServerCommandLine.java:149)
at org.apache.hadoop.hbase.master.HMaster.main(HMaster.java:3181)
Caused by: java.io.IOException: Failed update hbase:meta table descriptor
at org.apache.hadoop.hbase.util.FSTableDescriptors.tryUpdateMetaTableDescriptor(FSTableDescriptors.java:144)
at org.apache.hadoop.hbase.regionserver.HRegionServer.initializeFileSystem(HRegionServer.java:738)
at org.apache.hadoop.hbase.regionserver.HRegionServer.<init>(HRegionServer.java:635)
at org.apache.hadoop.hbase.master.HMaster.<init>(HMaster.java:528)
at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62)
at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
at java.lang.reflect.Constructor.newInstance(Constructor.java:423)
at org.apache.hadoop.hbase.master.HMaster.constructMaster(HMaster.java:3163)
... 5 more
查了很多资料，都没有解决。后来，在查看hdfs上也没有初始化的hbase文件夹。
排查原因：
这到底是什么情况，冷静下来后，继续看它的日志，日志说的很明白，就是没法更新他的元数据信息，hdfs都没有这个文件夹，它当然更新不了了，是不是这个元数据在创建的时候有问题？！
使用hadoop fs -mkdir /user/hbase，结果说没有权限不成功，我是用的是root用户没有权限穿件文件夹？？？我突然明白了，我虽然是root权限，但是对于Hadoop来说，如果它控制了权限，我依然创建不成功的。原因找到了，下面就是解决了。
解决方法：
hadoop的权限控制在hdfs-sit.xml中，这个我还是会的，于是，
1.进入/data/hadoop/hadoop/etc/hadoop 找到 hdfs-site.xml
2.vim hdfs-site.xml 看到果然dfs.permissions.enabled 没有加（默认是true）
3.加入
<property>
<name>dfs.permissions.enabled</name>
<value>false</value>
</property>
如果你的这个为true，改成false
:wq 保存退出
4.重启hadoop，先stop-yarn.sh，再stop-dfs.sh，然后jps看下是否停成功（查看datanode和namenode是否还存在），启动时，先启动start-dfs.sh，在start-yarn.xml
5.上一步启动成功后，就可以启动hbase了，这一次启动成功！
直接hbase shell （如果没有配置环境变量，则进入hbase的bin目录使用命令./hbase shell启动）如下图
