Although the Gmetad can be multi-layered, but the layer Gmetad all need to open gweb, or is very troublesome. If just worry about a gmetad unsafe, can be made into Gmetad high availability, but I do not know whether to think of Hadoop ha as automatic failover approach.
Resource arrangement:
Hdp1:gmetad, Gmond, Gweb
Hdp2:gmetad, Gmond, Gweb
Hdp3:gmond
Purpose of configuration:
HDP1 and HDP2 are Gmetad, gweb high availability, each node's gweb can show the entire cluster of data
HDP1, 2, 3 belong to a cluster
Configuration file:
Hdp3--gmond.conf:
Cluster {
Name = "HDP3"
Owner = "Unspecified"
Latlong = "Unspecified"
url = "Unspecified"
}
/* The host section describes attributes of the host, like the location */
Host {
Location = "Unspecified"
}
/* Feel free to specify as many udp_send_channels as. Gmond
Used to only support have a single channel */
Udp_send_channel {
#bind_hostname = yes # highly recommended, soon to is default.
# This option tells Gmond to use a source address
# That's resolves to the machine ' s hostname. Without
# This, the metrics could appear to come from any
# interface and the DNS names associated with
# those IPs would be used to create the rrds.
#mcast_join = 239.2.11.71
Host = Hdp1
Port = 8649
TTL = 1
}
Udp_send_channel {
#bind_hostname = yes # highly recommended, soon to is default.
# This option tells Gmond to use a source address
# That's resolves to the machine ' s hostname. Without
# This, the metrics could appear to come from any
# interface and the DNS names associated with
# those IPs would be used to create the rrds.
#mcast_join = 239.2.11.71
Host = HDP2
Port = 8649
TTL = 1
}
/* You can specify as many udp_recv_channels as. */
Udp_recv_channel {
#mcast_join = 239.2.11.71
#port = 8649
#bind = 239.2.11.71
#retry_bind = True
# Size of the UDP buffer. If You is handling lots of metrics you really
# should bump it up to e.g. 10MB or even higher.
# buffer = 10485760
}
/* You can specify as many tcp_accept_channels as a to share
An XML description of the state of the cluster * *
Tcp_accept_channel {
#port = 8649
# If you want to gzip XML output
Gzip_output = No
}
Hdp1--gmond.conf:
Cluster {
Name = "HDP3"
Owner = "Unspecified"
Latlong = "Unspecified"
url = "Unspecified"
}
/* The host section describes attributes of the host, like the location */
Host {
Location = "Unspecified"
}
/* Feel free to specify as many udp_send_channels as. Gmond
Used to only support have a single channel */
Udp_send_channel {
#bind_hostname = yes # highly recommended, soon to is default.
# This option tells Gmond to use a source address
# That's resolves to the machine ' s hostname. Without
# This, the metrics could appear to come from any
# interface and the DNS names associated with
# those IPs would be used to create the rrds.
#mcast_join = 239.2.11.71
Host = Hdp1
Port = 8649
TTL = 1
}
Udp_send_channel {
#bind_hostname = yes # highly recommended, soon to is default.
# This option tells Gmond to use a source address
# That's resolves to the machine ' s hostname. Without
# This, the metrics could appear to come from any
# interface and the DNS names associated with
# those IPs would be used to create the rrds.
#mcast_join = 239.2.11.71
Host = HDP2
Port = 8649
TTL = 1
}
/* You can specify as many udp_recv_channels as. */
Udp_recv_channel {
#mcast_join = 239.2.11.71
Port = 8649
#bind = 239.2.11.71
}
/* You can specify as many tcp_accept_channels as a to share
An XML description of the state of the cluster * *
Tcp_accept_channel {
Port = 8649
}
Hdp1--gmetad.conf:
[email protected] ganglia]# grep-v ^# gmetad.conf
Data_source "HDP3" localhost
RRAs "rra:average:0.5:1:244" "rra:average:0.5:24:244" "rra:average:0.5:168:244" "rra:average:0.5:672:244" \
gridname "HDP"
All_trusted on
Setuid_username Ganglia
Xml_port 8651
Interactive_port 8652
Case_sensitive_hostnames 0
Hdp2--gmond.conf:
Emulate HDP1 's configuration to
Effect achieved for configuration purposes
Single-layer Gmetad high availability