Hi, On Tue, Jan 20, 2009 at 5:04 PM, Ofer Inbar <[email protected]> wrote: > [snip] > > I think you'll find that if you left it at collect_once, restarted > your gmond's, and waited several hours, they'd eventually all get the > right data.
True - it took about 1h after restarting the gmonds on the nodes to finally converge to the correct number of total CPUs. > How many nodes do you have in a cluster, and how long after restarting > gmond's does your collect_once data converge? We typically have 2 node clusters of various types of machines. Thanks for the tip. Hope the developers find time to look at this. Gregory ------------------------------------------------------------------------------ This SF.net email is sponsored by: SourcForge Community SourceForge wants to tell your story. http://p.sf.net/sfu/sf-spreadtheword _______________________________________________ Ganglia-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ganglia-general
