Ambari Hive 无法正常查询 问题分析处理

1.今天遇到一个错误，在连接到hive之后，无法执行任何命令。

$ beeline -u "jdbc:hive2://hostname.domain.cn:10000/default;principal=hive/_HOST@domain.com;"Connecting to jdbc:hive2://hostname.domain.cn:10000/default;principal=hive/_HOST@domain.com;Connected to: Apache Hive (version release-1.2.1-EDH-1.1.2)Driver: Hive JDBC (version release-1.2.1-EDH-1.1.2)Transaction isolation: TRANSACTION_REPEATABLE_READBeeline version release-1.2.1-EDH-1.1.2 by Apache Hive0: jdbc:hive2://hostname.domain.cn:10000> show databases;

返回错误如下

Error: Error while processing statement: FAILED: Execution Error, return code 1 from org.apache.hadoop.hive.ql.exec.DDLTask. MetaException(message:For direct MetaStore DB connections, we don't support retries at the client level.) (state=08S01,code=1)

2. 百度这个错误，很多文档都提及是字符集的问题，需要保证mysql的字符集是latin1.

mysql>  show variables like '%collation_%';+----------------------+-------------------+| Variable_name        | Value             |+----------------------+-------------------+| collation_connection | latin1_swedish_ci || collation_database   | latin1_swedish_ci || collation_server     | latin1_swedish_ci |+----------------------+-------------------+

3. 如果字符集不是latin1，需要手动执行(本次问题，不是这个，我的字符集，原本就是latin1)

mysql>use hivemysql>alter database hive character set latin1;

4. 后来尝试重启一下hive，结果，metastore竟然启动失败。查看日志，说创建表失败。这个hive已经运行一段时间了，一直没有问题。对应的表也都是存在的，没有道理重建的。

5.日志输出如下：

Traceback (most recent call last):  File "/var/lib/ambari-agent/cache/common-services/HIVE/0.12.0.2.0/package/scripts/hive_metastore.py", line 259, in     HiveMetastore().execute()  File "/usr/lib/python2.6/site-packages/resource_management/libraries/script/script.py", line 280, in execute    method(env)  File "/var/lib/ambari-agent/cache/common-services/HIVE/0.12.0.2.0/package/scripts/hive_metastore.py", line 59, in start    self.configure(env)  File "/var/lib/ambari-agent/cache/common-services/HIVE/0.12.0.2.0/package/scripts/hive_metastore.py", line 73, in configure    hive(name = 'metastore')  File "/usr/lib/python2.6/site-packages/ambari_commons/os_family_impl.py", line 89, in thunk    return fn(*args, **kwargs)  File "/var/lib/ambari-agent/cache/common-services/HIVE/0.12.0.2.0/package/scripts/hive.py", line 320, in hive    user = params.hive_user  File "/usr/lib/python2.6/site-packages/resource_management/core/base.py", line 155, in __init__    self.env.run()  File "/usr/lib/python2.6/site-packages/resource_management/core/environment.py", line 160, in run    self.run_action(resource, action)  File "/usr/lib/python2.6/site-packages/resource_management/core/environment.py", line 124, in run_action    provider_action()  File "/usr/lib/python2.6/site-packages/resource_management/core/providers/system.py", line 273, in action_run    tries=self.resource.tries, try_sleep=self.resource.try_sleep)  File "/usr/lib/python2.6/site-packages/resource_management/core/shell.py", line 71, in inner    result = function(command, **kwargs)  File "/usr/lib/python2.6/site-packages/resource_management/core/shell.py", line 93, in checked_call    tries=tries, try_sleep=try_sleep)  File "/usr/lib/python2.6/site-packages/resource_management/core/shell.py", line 141, in _call_wrapper    result = _call(command, **kwargs_copy)  File "/usr/lib/python2.6/site-packages/resource_management/core/shell.py", line 294, in _call    raise Fail(err_msg)resource_management.core.exceptions.Fail: Execution of 'export HIVE_CONF_DIR=/usr/hdp/current/hive-metastore/conf/conf.server ; /usr/hdp/current/hive-metastore/bin/schematool -initSchema -dbType mysql -userName hive -passWord [PROTECTED] -verbose' returned 1. Metastore connection URL:     jdbc:mysql://hostname.mysql.cn/hivemetaMetastore Connection Driver :     com.mysql.jdbc.DriverMetastore connection User:     hiveStarting metastore schema initialization to 1.2.0Initialization script hive-schema-1.2.0.mysql.sqlConnecting to jdbc:mysql://hostname.mysql.cn/hivemetaConnected to: MySQL (version 5.1.73)Driver: MySQL Connector Java (version mysql-connector-java-5.1.40 ( Revision: 402933ef52cad9aa82624e80acbea46e3a701ce6 ))Transaction isolation: TRANSACTION_READ_COMMITTED0: jdbc:mysql://hostname.mysql.cn/hivem> !autocommit onAutocommit status: true0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40101 SET @OLD_CHARACTER_SET_C LIENT=@@CHARACTER_SET_CLIENT */No rows affected (0.003 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40101 SET @OLD_CHARACTER_SET_R ESULTS=@@CHARACTER_SET_RESULTS */No rows affected (0 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40101 SET @OLD_COLLATION_CONNE CTION=@@COLLATION_CONNECTION */No rows affected (0 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40101 SET NAMES utf8 */No rows affected (0 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40103 SET @OLD_TIME_ZONE=@@TIM E_ZONE */No rows affected (0.002 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40103 SET TIME_ZONE='+00:00' * /No rows affected (0 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40014 SET @OLD_UNIQUE_CHECKS=@ @UNIQUE_CHECKS, UNIQUE_CHECKS=0 */No rows affected (0 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40014 SET @OLD_FOREIGN_KEY_CHE CKS=@@FOREIGN_KEY_CHECKS, FOREIGN_KEY_CHECKS=0 */No rows affected (0.001 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40101 SET @OLD_SQL_MODE=@@SQL_ MODE, SQL_MODE='NO_AUTO_VALUE_ON_ZERO' */No rows affected (0 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40111 SET @OLD_SQL_NOTES=@@SQL _NOTES, SQL_NOTES=0 */No rows affected (0.001 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40101 SET @saved_cs_client      = @@character_set_client */No rows affected (0 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> /*!40101 SET character_set_client  = utf8 */No rows affected (0.001 seconds)0: jdbc:mysql://hostname.mysql.cn/hivem> CREATE TABLE IF NOT EXISTS `BUCKE TING_COLS` ( `SD_ID` bigint(20) NOT NULL, `BUCKET_COL_NAME` varchar(256) CHARACT ER SET latin1 COLLATE latin1_bin DEFAULT NULL, `INTEGER_IDX` int(11) NOT NULL, P RIMARY KEY (`SD_ID`,`INTEGER_IDX`), KEY `BUCKETING_COLS_N49` (`SD_ID`), CONSTRAI NT `BUCKETING_COLS_FK1` FOREIGN KEY (`SD_ID`) REFERENCES `SDS` (`SD_ID`) ) ENGIN E=InnoDB DEFAULT CHARSET=latin1Error: Can't create table 'hivemeta.bucketing_cols' (errno: 121) (state=HY000,code=1005)Closing: 0: jdbc:mysql://hostname.mysql.cn/hivemetaorg.apache.hadoop.hive.metastore.HiveMetaException: Schema initialization FAILED! Metastore state would be inconsistent !!org.apache.hadoop.hive.metastore.HiveMetaException: Schema initialization FAILED! Metastore state would be inconsistent !!    at org.apache.hive.beeline.HiveSchemaTool.doInit(HiveSchemaTool.java:270)    at org.apache.hive.beeline.HiveSchemaTool.doInit(HiveSchemaTool.java:243)    at org.apache.hive.beeline.HiveSchemaTool.main(HiveSchemaTool.java:473)    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)    at java.lang.reflect.Method.invoke(Method.java:606)    at org.apache.hadoop.util.RunJar.run(RunJar.java:221)    at org.apache.hadoop.util.RunJar.main(RunJar.java:136)Caused by: java.io.IOException: Schema script failed, errorcode 2    at org.apache.hive.beeline.HiveSchemaTool.runBeeLine(HiveSchemaTool.java:358)    at org.apache.hive.beeline.HiveSchemaTool.runBeeLine(HiveSchemaTool.java:326)    at org.apache.hive.beeline.HiveSchemaTool.doInit(HiveSchemaTool.java:266)    ... 8 more*** schemaTool failed ***

6.根据关键字"Schema initialization FAILED! Metastore state would be inconsistent !!"，查询到hdp的官方论坛。
解释说，这个是一个已知的问题。需要修改ambari的sql脚本，删除mysql的一些index等。
https://community.hortonworks.com/questions/113748/hive-metastore-start.html

7. 官方建议的步骤繁多。个人比较懒。而且这个hive库，前一天还是正常的。这个问题是突然出现的，我不倾向与，这个是bug的可能。
继续分析，查看hivemetastore的log，还是有必要的。
发现问题，表不存在

2018-03-07 15:34:43,871 INFO  [pool-8-thread-200]: metastore.HiveMetaStore (HiveMetaStore.java:logInfo(746)) - 200: get_all_databases2018-03-07 15:34:43,871 INFO  [pool-8-thread-200]: HiveMetaStore.audit (HiveMetaStore.java:logAuditEvent(371)) - ugi=ambari-qa-hbjt_hadoop@domain.com      ip=/172.18.98.57        cmd=get_all_databases2018-03-07 15:34:43,873 ERROR [pool-8-thread-200]: metastore.RetryingHMSHandler (RetryingHMSHandler.java:invoke(165)) - HMSHandler Fatal error: javax.jdo.JDOException: Exception thrown when executing query        at org.datanucleus.api.jdo.NucleusJDOHelper.getJDOExceptionForNucleusException(NucleusJDOHelper.java:596)      ......NestedThrowablesStackTrace:com.mysql.jdbc.exceptions.jdbc4.MySQLSyntaxErrorException: Table 'hivemeta.dbs' doesn't exist        at sun.reflect.GeneratedConstructorAccessor60.newInstance(Unknown Source)        at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)        at java.lang.reflect.Constructor.newInstance(Constructor.java:526)

8.现在要从mysql入手分析，逐步检查。

mysql> use hivemysql> show tables;+---------------------------+| Tables_in_hivemeta        |+---------------------------+| BUCKETING_COLS            || CDS                       || COLUMNS_V2                || COMPACTION_QUEUE          || COMPLETED_TXN_COMPONENTS  || DATABASE_PARAMS           || DBS                       || DB_PRIVS                  || DELEGATION_TOKENS         |

9. 问题来了，表不存在。为什么不存在呢？其实我已经知道问题了。Mysql表名大小写敏感问题。

mysql>select count(*) from `DBS`;ERROR 1146 (42S02): Table 'hivemeta.DBS' doesn't exist

10. 由于很多Oracle数据库的表名是大小写不敏感的。但是Mysql却是表名大小写敏感的。所以，很多开发人员，都喜欢将mysql设置为表名大小写不敏感。
而show tables之后，我就怀疑这个可能了。

11. 经过检查/etc/my.cnf，确实已经设置lower_case_table_names=1

12. Hive库创建的时候，mysql是大小写敏感的。所以，表名都是大写的。而后期，有人修改了mysql参数，取消大小写敏感。这样，问题就来了。Mysql并不是完全不敏感，而是在设置参数lower_case_table_names=1 之后，创建和查询的sql语句中的表名，默认修改为小写的。而已经创建的表名，还是大写的显示。但是你无论用大写表名，还是小写表名，都无法正确查询到了。

13. 这个问题出现的场景：
只有是在hive库创建完成之后，又修改/etc/my.cnf，添加参数lower_case_table_names=1导致的。

14. 问题找到之后，有两个解决方案：
a. 取消参数lower_case_table_names=1
b. 如果mysql的hive库中没有什么数据，删除hive库表，重新创建hive meta表，直接就创建出小写的表名了。
输出就是小写的了：

mysql> show tables;+---------------------------+| Tables_in_hive_beta       |+---------------------------+| bucketing_cols            || cds                       || columns_v2                || compaction_queue          || completed_txn_components  || database_params           || db_privs                  || dbs                       || delegation_tokens         || func_ru                   |

15. 我最后选择取消参数lower_case_table_names=1，并重启mysql。查询正常了。

mysql> select count(*) from `DBS`;+----------+| count(*) |+----------+|        2 |

16. Ambari启动hive metastore也成功了。
问题解决，一切恢复正常

$ beeline -u "jdbc:hive2://hostname.domain.cn:10000/default;principal=hive/_HOST@domain.com;"Connecting to jdbc:hive2://hostname.domain.cn:10000/default;principal=hive/_HOST@domain.com;Connected to: Apache Hive (version release-1.2.1-EDH-1.1.2)Driver: Hive JDBC (version release-1.2.1-EDH-1.1.2)Transaction isolation: TRANSACTION_REPEATABLE_READBeeline version release-1.2.1-EDH-1.1.2 by Apache Hive0: jdbc:hive2://hostname.domain.cn:10000> show databases;+----------------+--+| database_name  |+----------------+--+| default        || hbjt           |+----------------+--+2 rows selected (1.206 seconds)

最终的原因，就是有人修改的/etc/my.cnf，导致的本次问题。