Federico Sacerdoti <[EMAIL PROTECTED]> writes: > I have tagged CVS, built, and released Ganglia 2.5.1 on Sourceforge. > Please, please test it - the links, rpms, upgrade feature of the > webfrontend, etc.
OK, now I'm running it on 431 hosts in 8 clusters. Things seem to work nicely. The CPU usage seems fine - the context switch counter stays nice and low - and I can monitor the gmonds from Nagios without them dying the SIGPIPE death. Thanks, all! You might be interested in one of the uses I've put Ganglia to; on several clusters I extract batch queue stats from Maui and inject them into Ganglia using gmetric. Gmetad consolidates the data into rrds, from which I generate pretty graphs, published on: http://status.nsc.liu.se There are some gaps in the graphs where gmonds have died on me. I suspect the aforementioned SIGPIPE death. Let's see if things stabilize now. -- Leif Nixon Systems expert ------------------------------------------------------------ National Supercomputer Centre Linkoping University ------------------------------------------------------------