thanks! that should work alright then. > -----Original Message----- > From: Brad Nicholes [mailto:[EMAIL PROTECTED] > Sent: Wednesday, January 16, 2008 8:58 AM > To: [email protected]; Nordwall, Douglas J > Subject: Re: [Ganglia-general] modules and initiation > > >>> On 1/16/2008 at 9:52 AM, in message > <[EMAIL PROTECTED]>, Douglas > Nordwall <[EMAIL PROTECTED]> > wrote: > > are modules initiated at every collection time, or is it > initiated at > > ganglia start up and kept through the runtime? > > > > I'm asking because I am looking at doing per second metrics > based on > > the difference between the previously read value and the currently > > read value. The code (dstat, from the DAG repository) conveniently > > gives me nice classes for everything i'm looking at pulling out > > (interrupts per second, context switches per second, etc), and even > > goes so far as to check the value since the last time run, but of > > course, this produces a big spike in the second run and a 0 > value in > > the first (thus, giving you a spike). I figure I can > extract out the > > values a couple of times to get rid of the spike in the > instantiation > > and then go from there. > > > > The init_handler for the module is only called once at the > time when the module is loaded. The cleanup_handler is also > only called once just before the module is unloaded when > gmond is shutting down. The metric_handler itself is the > only function that is called for each collection interval. > > Brad > >
------------------------------------------------------------------------- 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
