Ganglia on Hadoop-2.7.1

Reference: https://www.digitalocean.com/community/tutorials/introduction-to-ganglia-on-ubuntu-14-04
OS: Ubuntu 14.04
Order

1. Gmetad install at master node (name node)
2. Gmond install all nodes which you want to know their source.(master + slaves)
3. synchronizing between hadoop and ganglia

To install on hadoop
Ganaglia version 3.6

• Gmetad (Ganglia meta daemon): a daemon on the master node that collects data from all the Gmond daemons (and other Gmetad daemons, if applicable).
• Gmond (Ganglia monitoring daemon): a small service that collects information about a node. This is installed on every server you want monitored.

1. Gmetad on master node

$sudo apt-get update
$sudo apt-get install -y ganglia-monitor rrdtool gmetad ganglia-webfrontend
Select yes(restart apache2) twice.
Copy  /etc/ganglia-webfrontend/apache.conf to apache sites-enabled to set up online graphical dashboard.
$sudo cp /etc/ganglia-webfrontend/apache.conf /etc/apache2/sites-enabled/ganglia.conf
Edit the Gmetad configuration file to set up your cluster.
$sudo nano /etc/ganglia/gmetad.conf
change:
data_source "my cluster" localhost -> data_source "Hadoop" 60 localhost
60 is option to measure every second.
Configure for Gmond on master noode.
$sudo vi /etc/ganglia/gmond.conf
change cluster name "my cluster" to "Hadoop"

And, set host at upd section.
udp_send_channel {
#mcast_join = 239.2.11.71 ## comment out
host = localhost
port = 8649
ttl = 1
}
udp_recv_channel {
#mcast_join = 239.2.11.71 ## comment out
port = 8649
#bind = 239.2.11.71 ## comment out
}
.
Restart
$sudo service ganglia-monitor restart && sudo service gmetad restart && sudo service apache2 restart
At browser
IP address/ganglia/

 
2. Gmond install all nodes which you want to know their source.(each slave nodes)
$sudo apt-get update
$sudo apt-get install -y ganglia-monitor
$sudo vi /etc/ganglia/gmond.conf
===============================================================
cluster {
name = "Hadoop" ## Cluster name
owner = "unspecified"
latlong = "unspecified"
url = "unspecified"
}
udp_send_channel {
#mcast_join = 239.2.11.71 ## Comment
host = 10.10.1.1 ## IP address of master node
port = 8649
ttl = 1
}
/* You can specify as many udp_recv_channels as you like as well.
udp_recv_channel {
mcast_join = 239.2.11.71
port = 8649
bind = 239.2.11.71
}
*/
============================================================
$sudo service ganglia-monitor restart
Set up Hadoop for monitoring
3. synchronizing between hadoop and ganglia (each slave nodes)
Just remove '#' and change servers ip address to master address ip
$sudo nano /usr/local/hadoop-2.7.1/etc/hadoop/hadoop-metrics2.properties
=========================================

*.sink.file.class=org.apache.hadoop.metrics2.sink.FileSink

*.sink.ganglia.class=org.apache.hadoop.metrics2.sink.ganglia.GangliaSink31

*.sink.ganglia.period=10

 

namenode.sink.ganglia.servers=master node IP address:8649

datanode.sink.ganglia.servers=master node IP address:8649

jobtracker.sink.ganglia.servers=master node IP address:8649

tasktracker.sink.ganglia.servers=master node IP address:8649

maptask.sink.ganglia.servers=master node IP address:8649

reducetask.sink.ganglia.servers=master node IP address:8649

 

==============================================

 

 

Individual Hadoop restart at each node
$sudo /usr/local/hadoop-2.7.1/sbin/hadoop-daemon.sh stop datanode
$sudo /usr/local/hadoop-2.7.1/sbin/hadoop-daemon.sh start datanode
Or, entire restart
$sudo /usr/local/hadoop-2.7.1/sbin/stop-all.sh
$sudo /usr/local/hadoop-2.7.1/sbin/start-all.sh

Restart
$sudo service ganglia-monitor restart && sudo service gmetad restart && sudo service apache2 restart