开启HadoopYarn的日志监控功能,配置Spark历史服务，解决web端查看日志时的Java.lang.ExceptionUnknown

解放方法

下来查询官方文档后，才了解到yarn的日志监控功能默认是处于关闭状态的，需要我们进行开启，开启步骤如下：

Ps:下面配置的文件的位置在hadoop根目录 etc/haddop文件夹下，比较老版本的Hadoop是在hadoop根目录下的conf文件夹中

本文hadoop配置环境目录：
/usr/local/src/hadoop-2.6.5/etc/hadoop

一、在yarn-site.xml文件中添加日志监控支持

<property>
    <name>yarn.log-aggregation-enable</name>
    <value>true</value>
</property>

二、在mapred-site.xml文件中添加日志服务的配置

<property>
    <!-- 表示提交到hadoop中的任务采用yarn来运行，要是已经有该配置则无需重复配置 -->
    <name>mapreduce.framework.name</name>
    <value>yarn</value>
</property>
<property>
  <!--日志监控服务的地址，一般填写为nodenode机器地址 -->
    <name>mapreduce.jobhistroy.address</name>
    <value>master:10020</value>
</property>
<property>
    <name>mapreduce.jobhistroy.webapp.address</name>
    <value>master:19888</value>
</property>

三、将修改后的配置文件拷贝到集群中的其他机器(单机版hadoop可以跳过该步骤)(也可以先不分发到各个节点上)

快捷一点可以使用 scp 命令将配置文件拷贝覆盖到其他机器

scp yarn-site.xml root@slave1:/usr/local/src/hadoop-2.6.5/etc/hadoop/

scp mapred-site.xml root@slave1:/usr/local/src/hadoop-2.6.5/etc/hadoop/

…其他datanode机器同理

四、配置spark

配置spark-defaults.conf文件
/usr/local/src/spark-2.4.4-bin-hadoop2.6/conf

vim spark-defaults.conf


#
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements.  See the NOTICE file distributed with
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#

# Default system properties included when running spark-submit.
# This is useful for setting default environmental settings.

# Example:
# spark.master                     spark://master:7077
# spark.eventLog.enabled           true
# spark.eventLog.dir               hdfs:/tmp/
# spark.serializer                 org.apache.spark.serializer.KryoSerializer
# spark.driver.memory              5g
# spark.executor.extraJavaOptions  -XX:+PrintGCDetails -Dkey=value -Dnumbers="one two three"



# 保存在本地
# spark.eventLog.dir=file://usr/local/hadoop-2.7.3/logs/
# spark.history.fs.logDirectory=file://usr/local/hadoop-2.7.3/logs/
#
spark.eventLog.enabled=true
spark.eventLog.compress=true
# 保存在hdfs上
spark.eventLog.dir=hdfs://master:9000/tmp/spark-yarn-logs
spark.history.fs.logDirectory=hdfs://master:9000/tmp/spark-yarn-logs
spark.yarn.historyServer.address=spark-master:18080

五、 创建HDFS 日志目录：

该目录与上述yarn-site.xml 中的目录需要一致

hdfs dfs -mkdir -p /tmp/spark-yarn-logs

六、上述配置完成后

1.重新启动hadoop

[root@master hadoop-2.6.5]# ./sbin/start-all.sh 

2.启动hadoop历史服务

mr-jobhistory-daemon.sh start historyserver

3.启动spark

在spark目录下

./sbin/start-all.sh

4.启动spark历史服务器

在spark目录下

sbin/start-history-server.sh

验证：
运行spark 在yarn模式PI

./bin/spark-submit --class org.apache.spark.examples.SparkPi --master yarn-cluster examples/jars/spark-examples_2.11-2.4.4.jar  10

5.关闭hadoop历史服务

mr-jobhistory-daemon.sh stop  historyserver

6.关闭spark历史服务

./sbin/stop-history-server.sh