记录ALiYun EMR常用服务的手动启动和停止命令(hdfs/yarn/mr-jobhistory/zk/spark-history)

最新推荐文章于 2024-09-07 00:17:30 发布

转载最新推荐文章于 2024-09-07 00:17:30 发布 · 422 阅读

0 ·

CC 4.0 BY-SA版权

原文链接：https://yq.aliyun.com/articles/62499

文章标签：

#大数据 #运维

本文档提供了详细的步骤说明如何使用root用户通过SSH登录到EMR集群的header节点，并执行一系列命令启动和调试ZooKeeper、HDFS、YARN及MapReduce History Server等组件。

说明：在ssh,用root用户登录到header(master)机器上,手工执行以下命令进行业务调试。

常用命令：

1. env (忘记是大写ENV,还是小写env,来查看环境变量)

2. cat /etc/hosts (查看EMR集群机器的hostname ip配置)

3. 在header机器上，切换到hadoop用户后，可以 ssh@emr-worker-1 直接登录到worker-1机器上，因为集群的机器，hadoop用户已经配置互相信任关系。

启动：

###start zk
su hadoop  -c "/usr/lib/zookeeper-current/bin/zkServer.sh start"
for i in `seq 1 2`;
do
 su hadoop -c "ssh emr-worker-${i} /usr/lib/zookeeper-current/bin/zkServer.sh start";
done

###start hdfs
su hdfs  -c "/usr/lib/hadoop-current/sbin/hadoop-daemon.sh start namenode"
su hdfs  -c "/usr/lib/hadoop-current/sbin/hadoop-daemon.sh start secondarynamenode"
for i in `seq 1 2`;
do
 su hdfs  -c  "ssh emr-worker-${i} /usr/lib/hadoop-current/sbin/hadoop-daemon.sh start datanode";
done

###start yarn
su hadoop  -c "/usr/lib/hadoop-current/sbin/yarn-daemon.sh start resourcemanager"
for i in `seq 1 2`;
do
 su hadoop -c "ssh emr-worker-${i} /usr/lib/hadoop-current/sbin/yarn-daemon.sh start nodemanager";
done

su hadoop  -c "/usr/lib/hadoop-current/sbin/yarn-daemon.sh start proxyserver"

###start mapreduce historyserver
su hadoop  -c "/usr/lib/hadoop-current/sbin/mr-jobhistory-daemon.sh start historyserver"

停止：

###stop spark
su hadoop  -c "/usr/lib/spark-current/sbin/stop-history-server.sh"

###stop mapreduce historyserver
su hadoop  -c "/usr/lib/hadoop-current/sbin/mr-jobhistory-daemon.sh stop historyserver"

###stop yarn
su hadoop  -c "/usr/lib/hadoop-current/sbin/yarn-daemon.sh stop proxyserver"

for i in `seq 1 2`;
do
 su hadoop -c "ssh emr-worker-${i} /usr/lib/hadoop-current/sbin/yarn-daemon.sh stop nodemanager";
done
su hadoop  -c "/usr/lib/hadoop-current/sbin/yarn-daemon.sh stop resourcemanager"

###stop hdfs
for i in `seq 1 2`;
do
 su hdfs  -c  "ssh emr-worker-${i} /usr/lib/hadoop-current/sbin/hadoop-daemon.sh stop datanode";
done
su hdfs  -c "/usr/lib/hadoop-current/sbin/hadoop-daemon.sh stop secondarynamenode"
su hdfs  -c "/usr/lib/hadoop-current/sbin/hadoop-daemon.sh stop namenode"

###stop zk
su hadoop  -c "/usr/lib/zookeeper-current/bin/zkServer.sh stop"
for i in `seq 1 2`;
do
 su hadoop -c "ssh emr-worker-${i} /usr/lib/zookeeper-current/bin/zkServer.sh stop";
done