This paper derives from:Everyday Learning IT - Knowledge Base

Cluster machine

Create management user hadoop (all node operations)

useradd hadoop

Modify the machine hosts file (all node operations)

echo -e '192.168.99.181 ambari-mirror\n192.168.99.101 ambari-server\n192.168.99.106 ambari-agent1\n192.168.99.107 ambari-agent2' >> /etc/hosts

Set hadoop user sudo password-free (all node operations)

Use visudo to add after line 99

hadoop ALL=(ALL) NOPASSWD: ALL

Configure hadoop user keyless login (operated on ambari-mirror node)

Enter hadoop users

su hadoop

Generate a key (just go all the way back)

ssh-keygen

Added to authorized keys

cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys

Modify key permissions

chmod 700 ~/.ssh
chmod 600 ~/.ssh/authorized_keys

Copy keys to other nodes (using root users, operating on ambari-mirror nodes)

sudo scp -r /home/hadoop/.ssh/ root@ambari-server:/home/hadoop/
sudo scp -r /home/hadoop/.ssh/ root@ambari-agent1:/home/hadoop/
sudo scp -r /home/hadoop/.ssh/ root@ambari-agent2:/home/hadoop/

NTP service (operates on all nodes)

Installation service: sudo yum install ntp
 Set boot start: sudo system CTL enable ntpd
 Start service: sudo system CTL start ntpd

Firewall (operated on all nodes)

Close the firewall: sudo service firewalld stop

Close selinux (operates on all nodes)

Use / usr/sbin/sestatus -v to check if selinux is enable closed

sudo sed -i 's/SELINUX=enforcing/SELINUX=disabled/g' /etc/selinux/config

Restart your computer to enter hadoop users using sudo reboot

Setting up UMASK (using root user, operating on all nodes)

echo umask 0022 >> /etc/profile && source /etc/profile

Install dependency packages (using hadoop users, operating on ambari-mirror nodes)

sudo yum install yum-utils createrepo httpd yum-plugin-priorities

Setting up pluginconf (using root user, operating on ambari-mirror node)

echo 'gpgcheck=0' >> /etc/yum/pluginconf.d/priorities.conf

Download the ambari installation source (using hadoop users, operating on the ambari-mirror node)

sudo wget -nv http://public-repo-1.hortonworks.com/ambari/centos7/2.x/updates/2.2.2.0/ambari.repo  -O /etc/yum.repos.d/ambari.repo
sudo wget -nv http://public-repo-1.hortonworks.com/HDP/centos7/2.x/updates/2.4.2.0/hdp.repo  -O /etc/yum.repos.d/HDP.repo

Check the availability of the installation source (using hadoop users, operating on the ambari-mirror node)

yum repolist

Ensure that you have hdp sources

Create centos7 folder (using hadoop users, operating on ambari-mirror nodes)

sudo mkdir -p ambari/centos7 && cd /var/www/html/ambari/centos7/

Synchronize ambari source to local (using hadoop users, operating on ambari-mirror nodes)

sudo reposync -r Updates-ambari-2.2.2.0
sudo reposync -r HDP-2.4.2.0
sudo reposync -r HDP-UTILS-1.1.0.20

Generate local warehouse source data (using hadoop users, operating on ambari-mirror nodes)

Create ambari metadata

sudo createrepo /var/www/html/ambari/centos7/Updates-ambari-2.2.2.0/

Spawning worker 0 with 8 pkgs
Workers Finished
Saving Primary metadata
Saving file lists metadata
Saving other metadata
Generating sqlite DBs
Sqlite DBs complete

Create hdp metadata

sudo createrepo /var/www/html/hdp/centos7/2.4.X/HDP-2.4.2.0/

Spawning worker 0 with 181 pkgs
Workers Finished
Saving Primary metadata
Saving file lists metadata
Saving other metadata
Generating sqlite DBs
Sqlite DBs complete

Create hdp util metadata

shellsudo createrepo /var/www/html/hdp/centos7/2.4.X/HDP-UTILS-1.1.0.20/ 

Spawning worker 0 with 44 pkgs
Workers Finished
Saving Primary metadata
Saving file lists metadata
Saving other metadata
Generating sqlite DBs
Sqlite DBs complete

Start the http service (using hadoop users, operating on ambari-mirror nodes)

sudo service httpd start

View Ambari source availability (using hadoop users, operating on ambari-mirror nodes)

http://ambari-mirror/ambari/
http://ambari-mirror/hdp/