|
Problem : The installation was pretty smooth using yum but I have a problem with connectivity between gmond and gmetad. I have a cluster to monitor, made up of three virtual machines (viz. ClusterM, node1, node2) having centOS 5. All three of them are having gmond running and gmetad is running on clusterM which is head node. But Meta node is able to see(on web) only local gmond host for clusterM only (itself). I have tried the "telnet localhost 8694" to check the XML on all nodes and find that its generating fine then tried the "telnet localhost 8651" on meta node for gmetad but its showing the XML for itself only. Details are as follows : clusterM : 192.168.122.30 node1 : 192.168.122.82 node2 : 192.168.122.77 The configuration of gmetad (on clusterM 192.168.122.30): data_source "my cluster" 300 192.168.122.30:8649 192.168.122.82:8649 192.168.122.77:8649 The configuration of gmond(on clusterM, node1 & node2): cluster { name = "my cluster" 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 you like. Gmond used to only support having a single channel / udp_send_channel { #bind_hostname = yes # Highly recommended, soon to be default. # This option tells gmond to use a source address # that resolves to the machine's hostname. Without # this, the metrics may appear to come from any # interface and the DNS names associated with # those IPs will be used to create the RRDs. mcast_join = 239.2.11.71 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 } / You can specify as many tcp_accept_channels as you like to share an xml description of the state of the cluster / tcp_accept_channel { port = 8649 } Coz i m a beginner in this field I've tried various configuration options for both gmond & gmetad available on web but none of them worked for me..! What I'm missing here? post a reply or mail me at gmail.com">anilbhargava777@gmail.com Many Thanx in advance...:) Submitted By : Anil Bhargava anilbhargava777@gmail.com |
|
your gmond.conf (on each node that you are monitoring) should have something like udp_send_channel { |
|
Thanks a lot for ur answer but this configuration is also not working. Then I mentioned 3 data sources in gmetad config file for all 3 of my vm's then it is showing the graphs in browser by treating all of them as a seperate clusters. But i want all 3 of them to be part of a single cluster. I wanted to ask you one thing about your rply that has the line host = "<ip of="" the="" monitoring="" head="" node="">" so should it be the head node's IP in gmond of all nodes or should it be their respective IP's. Thanx in advance...:) |
|
Sorry, I was giving you config info for Ganglia running in unicast mode. I see now that your config is for Ganglia running in multicast mode. I haven't tried multicast mode myself, but different sources I've read say the same thing about the issue of getting only stats for the head node when running in multicast mode -> it's due to multicast protocol being blocked by the network hardware between the server node and the rest of the cluster; check the switch configuration to make sure that it allows multicast. In the meantime, try unicast mode by changing the udp_send_channel entry (in gmond.conf) , FROM udp_send_channel {mcast_join = 239.2.11.71 port = 8649 ttl = 1 } TO udp_send_channel {bind_hostname = yes host = HEAD_NODE_IP port = 8649 ttl = 1 } |
