Hi list:

I am currently using Ganglia v2.5.6 and am having a strange behaviour:

I have 3 data_sources, for three different types of cluster.  Each
listens to their own multicast port (mcast_port).

Data_source A has 13 nodes, and when I add or remove nodes from
data_source B, data_source A would have all but 1 node down (the one
remaining would be the one I have put down in gmetad.conf).  After a few
minutes data_source A would become back to normal showing all 13 nodes
being up.

Data_source A uses mcast_port 8650 and data_source B uses mcast_port
8652.

If you have any idea what is causing the problem, please let me know.

Thanks,

Bernard

Reply via email to