1)ganglia
gmond and gmetad it will communicate via multicast IP.
So
we have to configure multicast IP as route in eth0 (also private IP
also should configure in eth0)
/sbin/route
add -host 239.2.11.71 dev eth0
Note:1)Multicast
IP Range From 224.0.0.0 To 239.255.255.255.
2)Configuration
(/etc/gmond.conf)
udp_recv_channel
{
mcast_join = 239.2.11.71
port = 8649
bind = 239.2.11.71
}
3)Verification
method.
#
netstat -tulpn | grep gmond
tcp
0 0 0.0.0.0:8649 0.0.0.0:*
LISTEN 4946/gmond
udp
0 0 239.2.11.71:8649 0.0.0.0:*
4946/gmond
Useful
command.
1)#
gstat -d 5
CLUSTER
INFORMATION
Name: HCL Cluster
Hosts: 1 [ No Of the
Hosts are interacting]
Gexec
Hosts: 0
Dead Hosts: 0
Localtime: Mon Feb 4
16:48:33 2013
DEAD
CLUSTER HOSTS
There
are no hosts down at this time
2)#
gmond -f -d 10 [ -f run in foreground -d for debug level 10 ]
loaded
module: core_metrics
loaded
module: cpu_module
loaded
module: disk_module
loaded
module: load_module
loaded
module: mem_module
loaded
module: net_module
loaded
module: proc_module
loaded
module: sys_module
loaded
module: multicpu_module
loaded
module: python_module
Discovered
device /
Discovered
device /opt
Discovered
device /boot
udp_recv_channel
mcast_join=239.2.11.71 mcast_if=NULL port=8649 bind=239.2.11.71
tcp_accept_channel
bind=NULL port=8649
Unable
to create tcp_accept_channel. Exiting.
3)Ganglia
Communication.
[root@master ~]# netstat -tulp | egrep 'gmond|gmetad'
tcp 0 0 *:8649 *:* LISTEN 5667/gmond
tcp 0 0 *:8651 *:* LISTEN 5677/gmetad
tcp 0 0 *:8652 *:* LISTEN 5677/gmetad
udp 0 0 239.2.11.71:8649 *:* 5667/gmond
[root@node3 ~]# netstat -tulpn | grep gmond
tcp 0 0 0.0.0.0:8649 0.0.0.0:* LISTEN 4946/gmond
udp 0 0 239.2.11.71:8649 0.0.0.0:* 4946/gmond
tcp 0 0 *:8649 *:* LISTEN 5667/gmond
tcp 0 0 *:8651 *:* LISTEN 5677/gmetad
tcp 0 0 *:8652 *:* LISTEN 5677/gmetad
udp 0 0 239.2.11.71:8649 *:* 5667/gmond
[root@node3 ~]# netstat -tulpn | grep gmond
tcp 0 0 0.0.0.0:8649 0.0.0.0:* LISTEN 4946/gmond
udp 0 0 239.2.11.71:8649 0.0.0.0:* 4946/gmond
Common
Error Message
ERROR
1: If we are not properly configured the multicast IP as router
then we will get the following error message.
1)/var/log/message
Feb
1 17:07:45 master /usr/sbin/gmond[11378]: Unable to create UDP
client for master@hcl.com:8649. Exiting.
Feb
1 17:07:46 master /usr/sbin/gmetad[11333]: data_thread() got no
answer from any [IIT_GUWAHATI] datasource
Above
error we were getting due to no communication between gmond and
gmetad on sameserve
Error
2: # /etc/init.d/gmond status
gmond
dead but subsys locked
Solution: configure mulitcast IP as router.
Note
: In the master Node. We should configure.
Eth0
as the private IP(10.0.0.0)
Eth1
should configure as Public IP(192.168.0.0) to maintain the proper communication between the nodes and master.
0 comments:
Post a Comment