执行spark sql 遇到的问题

运行环境：

用图形更直观点。

在 spark cluster 和 yarn cluster 两种方式运行spark sql， 操作hive中的数据，另外，hive 是独立的，可以直接运行hive处理数据。

spark sql的程序比较好写，直接看spark的example的例子HiveFromSpark ，很容易理解

首先，在spark cluster上运行：

将hive的   hive-site.xml  配置文件放到 ${SPARK_HOME}/conf 目录下

1. #!/bin/bash
3. cd $SPARK_HOME
4. ./bin/spark-submit \
5. -- class com.datateam.spark.sql.HotelHive \
6. --master spark://192.168.44.80:8070 \
7. --executor-memory 2G \
8. --total-executor-cores 10 \
9. /home/q/spark/spark-1.1.1-SNAPSHOT-bin-2.2.0/jobs/spark-jobs-20141023.jar \

执行脚本，遇到下面的错误：

1. Exception in thread "main" org.apache.hadoop.hive.ql.metadata.HiveException: Unable to fetch table dw_hotel_price_log
2. at org.apache.hadoop.hive.ql.metadata.Hive.getTable(Hive.java: 958)
3. at org.apache.hadoop.hive.ql.metadata.Hive.getTable(Hive.java: 924)
4. ……
5. Caused by: org.datanucleus.exceptions.NucleusException: Attempt to invoke the "BONECP" plugin to create a ConnectionPool gave an error :
6. The specified datastore driver ("com.mysql.jdbc.Driver") was not found in the CLASSPATH.
7. Please check your CLASSPATH specification, and the name of the driver.
8. at org.datanucleus.store.rdbms.ConnectionFactoryImpl.generateDataSources(ConnectionFactoryImpl.java:237)
9. at org.datanucleus.store.rdbms.ConnectionFactoryImpl.initialiseDataSources(ConnectionFactoryImpl.java:110)
10. at org.datanucleus.store.rdbms.ConnectionFactoryImpl.<init>(ConnectionFactoryImpl.java:82)
11. ... 127 more
12. Caused by: org.datanucleus.store.rdbms.datasource.DatastoreDriverNotFoundException: The specified datastore driver ("com.mysql.jdbc.Driver") was not
13.  found in the CLASSPATH. Please check your CLASSPATH specification, and the name of the driver.
14. at org.datanucleus.store.rdbms.datasource.AbstractDataSourceFactory.loadDriver(AbstractDataSourceFactory.java:58)
15. at org.datanucleus.store.rdbms.datasource.BoneCPDataSourceFactory.makePooledDataSource(BoneCPDataSourceFactory.java:61)
16. at org.datanucleus.store.rdbms.ConnectionFactoryImpl.generateDataSources(ConnectionFactoryImpl.java:217)

意思是找不到 jdbc 的 connector，解决办法：

在提交任务的脚本里加入下面的配置语句即可：

    --driver-class-path /home/q/spark/spark-1.1.1-SNAPSHOT-bin-2.2.0/lib/mysql-connector-java-5.1.22-bin.jar \

spark cluster 遇到的问题不多，主要在yarn cluster上遇到几个问题。

在spark cluster上调用hive的数据，需要将 hive-site.xml 文件放到spark的conf 目录下，那在yarn上运行该将hive的配置文件放到哪里才能被 spark sql 识别呢？

在提交任务的时候加上：

--files /home/q/spark/spark-1.1.1-SNAPSHOT-bin-2.2.0/conf/hive-site.xml \

这里用到的是 --files  ， 而不是 --conf

先看一下 提交任务的脚本：

1. cd $SPARK_HOME
2. ./bin/spark-submit -- class com.qunar.datateam.spark.sql.HotelHive \
3. --master yarn-cluster \
4. --num-executors 10 \
5. --driver-memory 4g \
6. --executor-memory 2g \
7. --executor-cores 2 \
8. --files /home/q/spark/spark-1.1.1-SNAPSHOT-bin-2.2.0/conf/hive-site.xml \
9. /home/q/spark/spark-1.1.1-SNAPSHOT-bin-2.2.0/jobs/spark-jobs-20141023.jar \

ok，我们这里同样需要将mysql connector的jar包添加进去，如何进行？

--jars mysql-connectorpath

但是会出现下面的问题：

1. Exception in thread "Driver" java.lang.reflect.InvocationTargetException
2. at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
3. at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java: 57)
4. at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java: 43)
5. at java.lang.reflect.Method.invoke(Method.java: 606)
6. at org.apache.spark.deploy.yarn.ApplicationMaster$$anon$ 2.run(ApplicationMaster.scala: 162)
7. ……
9.  Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: Unable to fetch table tablename
10.  at org.apache.hadoop.hive.ql.metadata.Hive.getTable(Hive.java: 958)
11. at org.apache.hadoop.hive.ql.metadata.Hive.getTable(Hive.java: 924)
12. Caused by: java.lang.RuntimeException: Unable to instantiate org.apache.hadoop.hive.metastore.HiveMetaStoreClient
13.         at org.apache.hadoop.hive.metastore.MetaStoreUtils.newInstance(MetaStoreUtils.java: 1212)
14.         at org.apache.hadoop.hive.metastore.RetryingMetaStoreClient.<init>(RetryingMetaStoreClient.java: 62)
15.         at org.apache.hadoop.hive.metastore.RetryingMetaStoreClient.getProxy(RetryingMetaStoreClient.java: 72)
16.         at org.apache.hadoop.hive.ql.metadata.Hive.createMetaStoreClient(Hive.java: 2372)
17.         at org.apache.hadoop.hive.ql.metadata.Hive.getMSC(Hive.java: 2383)
18.         at org.apache.hadoop.hive.ql.metadata.Hive.getTable(Hive.java: 950)
19.         ... 68 more
20. Caused by: java.lang.reflect.InvocationTargetException
21.         at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
22.         at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java: 57)
23.         at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java: 45)
24.         at java.lang.reflect.Constructor.newInstance(Constructor.java: 526)
25.         at org.apache.hadoop.hive.metastore.MetaStoreUtils.newInstance(MetaStoreUtils.java: 1210)
26.         ... 73 more
27. Caused by: javax.jdo.JDOFatalUserException: Class org.datanucleus.api.jdo.JDOPersistenceManagerFactory was not found.
28. NestedThrowables:
29. java.lang.ClassNotFoundException: org.datanucleus.api.jdo.JDOPersistenceManagerFactory
30. ……
31. Caused by: java.lang.ClassNotFoundException: org.datanucleus.api.jdo.JDOPersistenceManagerFactory
32.         at java.net.URLClassLoader$ 1.run(URLClassLoader.java: 366)
33.         at java.net.URLClassLoader$ 1.run(URLClassLoader.java: 355)
34.         at java.security.AccessController.doPrivileged(Native Method)
35.         at java.net.URLClassLoader.findClass(URLClassLoader.java: 354)
36.         at java.lang.ClassLoader.loadClass(ClassLoader.java: 425)
37.         at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java: 308)
38.         at java.lang.ClassLoader.loadClass(ClassLoader.java: 358)
39.         at java.lang.Class.forName0(Native Method)
40.         at java.lang.Class.forName(Class.java: 270)
41.         at javax.jdo.JDOHelper$ 18.run(JDOHelper.java: 2018)
42.         at javax.jdo.JDOHelper$ 18.run(JDOHelper.java: 2016)
43.         at java.security.AccessController.doPrivileged(Native Method)
44.         at javax.jdo.JDOHelper.forName(JDOHelper.java: 2015)
45.         at javax.jdo.JDOHelper.invokeGetPersistenceManagerFactoryOnImplementation(JDOHelper.java: 1162)
46.         ... 97 more

查找资料，发现别人的讨论：http://apache-spark-user-list.1001560.n3.nabble.com/Spark-SQL-JDBC-td11369.html

于是将

datanucleus-api-jdo-3.2.1.jar, datanucleus-core-3.2.2.jar, datanucleus-rdbms-3.2.1.jar 

都加到 --jars 里，但是还是出问题：

1. Exception in thread "Driver" java.lang.reflect.InvocationTargetException
2. at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
3. at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java: 57)
4. at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java: 43)
5. at java.lang.reflect.Method.invoke(Method.java: 606)
6. at org.apache.spark.deploy.yarn.ApplicationMaster$$anon$ 2.run(ApplicationMaster.scala: 162)
7. Caused by: org.apache.spark.SparkException: Job aborted due to stage failure: Task 6 in stage 2.0 failed 4 times, most recent failure:
8. Lost task 6.3 in stage 2.0 (TID 34, l-hbase72.data.cn8
9. ): java.io.FileNotFoundException: ./datanucleus-core- 3.2.2.jar (Permission denied)
10. java.io.FileOutputStream.open(Native Method)
11. java.io.FileOutputStream.<init>(FileOutputStream.java: 221)
12. com.google.common.io.Files$FileByteSink.openStream(Files.java: 223)
13. com.google.common.io.Files$FileByteSink.openStream(Files.java: 211)
14. com.google.common.io.ByteSource.copyTo(ByteSource.java: 203)
15. com.google.common.io.Files.copy(Files.java: 436)

经过不断尝试，将 --jars 后面的配置的jar包都用 --archives 的方式打到运行jar中:

--archives mysql-connector.jar,datanucleus-api-jdo-3.2.1.jar, datanucleus-core-3.2.2.jar, datanucleus-rdbms-3.2.1.jar 

ok

另外还要注意一点：

spark sql 中不认“；”，所以只能在sql中指明database，不能用 use database ； 这样的hive sql 语句指定database