Shulou(Shulou.com)06/01 Report--

When you do not modify the PID file location of HADOOP/HBASE/SPARK, the system will generate the PID file to the / tmp directory by default, but the / tmp directory will be deleted after a period of time, so later when we stop HADOOP/HBASE/SPARK, we will find that we can not stop the corresponding process, because the PID file has been deleted, so now we need to change the PID file address of HADOOP/HBASE/SPARK. The modification method is as follows:

We need to generate the PID file of HADOOP/HBASE/SPARK into the / var/hadoop/pids directory.

Create a new / var/hadoop/pids directory on each node of the cluster command: sudo mkdir-p / var/hadoop/pids

I need to modify the permissions of the / var/hadoop/pids directory on each node. I use the hadoop user to start the HADOOP/HBASE/SPARK cluster, so I need to change the directory permissions of / var/hadoop/pids to hadoop, otherwise I cannot write to this directory when I start the HADOOP/HBASE/SPARK cluster. Modify permission command: chown-R hadoop:hadoop / var/hadoop

Modify hadoop-env.sh to add: export HADOOP_PID_DIR=/var/hadoop/pids

Modify yarn-env.sh to add: export YARN_PID_DIR=/var/hadoop/pids

Modify hbase-env.sh and add: export HBASE_PID_DIR=/var/hadoop/pids

Modify spark-env.sh and add: export SPARK_PID_DIR=/var/hadoop/pids

Send the configuration file to each node in the cluster, restart the cluster, and look at the / var/hadoop/pids directory to find that the PID file is generated in that directory.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.