HADOOP cluster installation
• Use a browser to enter the configuration page of Cloudera Manager , the address is http:// host node IP:7180 , the default username / password is admin/admin
• After logging in, enter the CM page, check YES in the red box , and then click Continue
• After selecting the free version of Cloudera Express , click Continue
• Click to continue
• Select the currently managed host
• Select all hosts and click Continue
• Go to the Install CDH Options page .
If there is no CDH version to choose from, click More Options, and in "Remote Parcel Repository URL", add a http://192.168.186.109/cdh55/parcels/5.5.0 (note that the red font is the host that needs to be modified node ip address); if there is a choice, select the corresponding version.
• Click Continue to enter the Install JDK page. Since JDK has been manually installed before , click Continue directly in this step
• Enter the cluster installation page and wait for the installation to complete
• After the installation is complete, it will turn to the following page, click Continue
• Enter the host correctness check page. After the check result is completed, if some configurations are abnormal, modify them according to the prompts. The errors encountered in this installation are as follows:
a) Set /proc/sys/vm/swappiness to 0 and execute the following command to modify
sysctl vm.swappiness=0
echo 'vm.swappiness=0'>> /etc/sysctl.conf
b) "Transparent Huge Pages" is enabled, it may cause major performance issues, execute the following command to modify
echo never > /sys/kernel/mm/transparent_hugepage/defrag
echo "echo never > /sys/kernel/mm/transparent_hugepage/defrag" >> /etc/rc.local
echo never > /sys/kernel/mm/transparent_hugepage/enabled
echo "echo never > /sys/kernel/mm/transparent_hugepage/enabled" >> /etc/rc.local
• Check again after modification, and click Finish after all items are normal
• Go to the cluster settings page
• Select the " Kernel with Spark " option and click Continue
• Use default values (separate components into different nodes if necessary), click Continue
• Enter the database configuration page, you need to skip to Chapter 4 to configure the MariaDB database before configuring the database.
• Enter the information about the hive and Oozie databases created in MariaDB before, click " Test Connection " , and click Continue when finished
• Go to the Waiting for Review Changes page and wait for the review changes to complete
• After reviewing the changes, click Continue
• Start installing Hadoop components and wait for completion
• After the installation is complete, click Continue
• To complete the installation, click Finish
• Click Finish to enter the Cloudera Manager page