Spark to yarn启动的时候报错：org.apache.spark.SparkException: Yarn application has already ended!

场景：Spark to yarn启动的时候报错
原因：虚拟内存不够
解决方案：

                  可通过以下两种方式解决：

                 调大虚拟内存比：

                              yarn.nodemanager.vmem-pmem-ratio 默认2.1倍

                 或者关闭虚拟内存检查：

                             yarn.nodemanager.vmem-check-enabled 默认true

这是犯错的场景：

输入命令如下： 

./spark-submit  --master yarn --class org.apache.spark.examples.SparkPi ../examples/jars/spark-examples_2.11-2.2.1.jar 1

报错如下：

org.apache.spark.SparkException: Yarn application has already ended! It might have been killed or unable to launch application master.

打开hadoop的web端

解决方案如下：

切换到hadoop目录下

cd /home/hadoop-2.6.5/etc/hadoop

vim yarn-site.xml

<property>
<name>yarn.nodemanager.vmem-pmem-ratio</name>
<value>3</value>
</property>

重启yarn

stop-yarn.sh

start-yarn.sh

重新输入指令

./spark-submit  --master yarn --class org.apache.spark.examples.SparkPi ../examples/jars/spark-examples_2.11-2.2.1.jar 1