Hi Nick. I went hunting around the developer list, and someone had already submitted a patch (#143) for RRD COUNTER support. It seems to work fine on my test bed.
Unfortunately the derive case isn't as easy as I thought. regards (and thanks to everyone for their ideas) Ian Nick Galbreath wrote: > > > > Howdy, > > For Derive types I have no idea. But for counters... > > First, the 3.0.4 version of gmetric/gmond has a bug that doesn't allow > you to send "slope positive" (i..e always increasing == counter). If > you pull mainline, it is fixed. > > Second, I have a patch in bugzilla (with a typo, I need to update) > that allows one to use gmetric to add COUNTER data. (i.e. if gmetric > send in a slope positive, then make a COUNTER). > > BUT.. another way of doing this, I just thought, is to create the rrd > file BEFORE you start adding the data. (or if you have the RRD > already, zap it, and redo). Once the RRD file is made, gmetad does > not care and just adds the data. > > The trick is duplicating what gmetad does in creating the file. This > might be an easy utility to whip up.. might need a bit of restructing > of the code, but the patch should be minimal. > > I hope to look at all this again over the weekend. > > --nickg > ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Ganglia-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ganglia-general
