Hi Jonathan: On Fri, Jul 16, 2010 at 12:24 PM, Jonathan Weiss <[email protected]> wrote:
>> - run the gmonds on each "cluster" on a dedicated port (per cluster) >> - let them cast their messages to a dedicated "aggregator" gmond for each >> cluster >> - let gmetad query those "aggregators" on their dedicated ports > > This is what I feared. Is there no way to have gmonds from different > cluster report to the same gmond? > I don't want to have a separated aggregator node per role/cluster as > my nodes come and go in EC2. In my case all instances on EC2 are > dynamic and scale up and down on demand, so I would have to query all > gmonds in a role. As Martin mentioned, you could run multiple gmonds with different ports on the same host. Alternatively, you can also specify multiple udp_recv_channels using different ports. I haven't tried this in practice, but should do what you want. > Is this a feasible way? To have gmetad query maybe ~50 gmonds that are > grouped in around 5-6 clusters? This only works if each gmond sends its metric to every other gmond, in essence re-creating the behaviour of multicast. Cheers, Bernard ------------------------------------------------------------------------------ This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first _______________________________________________ Ganglia-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ganglia-general
