>> I'm trying to configure ganglia (3.0.7) to monitor 120 nodes. It works >> fine if I just enter a small number of nodes as data_source in the >> gmetad.conf file, just like all the documentation shows. But if I try to >> enter "too many" nodes, gmetad segfaults at startup.
I noticed that too. So I hacked my config-file generator to only list a dozen random nodes instead of all of them. >Is there a reason why you would want to list all 120 nodes in the >data_source directive of gmetad? I was doing it to maximize redundancy. All my nodes are multicasting, so I consider them all equal peers; there is no "head" node in the cluster. Gmetad should therefore be able to talk to any one of them and get the same result with no single point of failure. In fact, I don't try the nodes in the order listed either. I use the patch from bz#92 to talk to a random node since I didn't like a bad gmond taking down an entire cluster when the rest of the cluster is actually up and running. http://bugzilla.ganglia.info/cgi-bin/bugzilla/show_bug.cgi?id=92 -twitham ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Ganglia-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ganglia-general
