1. 抛出异常,准确的捕获异常,并输出。运行时间 : 7mins, 47sec,输出数据 64.87M
try{
}catch{
case remoteException: RemoteException => println("ERROR++++++org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.hdfs.protocol.RecoveryInProgressException): Failed to APPEND_FILE /tmp/data/cxb/ecar_null_photo_unmatch/ds=2020xxxx/unmatchxx.log for DFSClient_NONMAPREDUCE_385164413_56 on 172.16.1.53 because another recovery is in progress by DFSClient_NONMAPREDUCE_-2142824852_51 on 172.16.1.53")
case exception: Exception => exception.printStackTrace()
}
2. 抛出异常,没有准确的捕获异常,没有输出。运行时间 : 5mins, 59sec,输出数据:51.75M
try{
}catch{
case _ => "ERROR++++++org.apache.hadoop.ipc.RemoteException"
}
3.抛出异常,准确的捕获异常,没有输出。运行时间 : 5mins, 50sec,输出数据:49.55M
try{
}catch{
case remoteException: RemoteException => "ERROR++++++org.apache.hadoop.ipc.RemoteException"
case exception: Exception => exception.printStackTrace()
}
以上的资源情况:
--driver-memory 1G --executor-memory 6G --executor-cores 2 --num-executors 10
1.修改资源,用上面的异常3运行, 运行时间 : 6mins, 14sec,输出数据:47M
--driver-memory 1G --executor-memory 6G --executor-cores 2
2.再次修改资源 ,只设置了8G内存,其他的都没设置
--executor-memory 8G
发现只有两个executor在运行
3.再次改资源 ,7mins, 24sec,60.44M
--driver-memory 1G --executor-memory 8G --executor-cores 2 --num-executors 5
4.再次改资源 , 6mins, 45sec,57.75M
--executor-memory 1G --num-executors 10
5.再改 5mins, 33sec 输出47M
--executor-memory 4G --num-executors 10
6.改 3mins, 29sec ,输出27M (第二次 3mins, 17sec ,25M)
--executor-memory 4G --executor-cores 2 --num-executors 10