On Saturday 10 April 2004 12:11 am, Tom Shilson wrote: > Can you give us some details? Was it the toolkit reporting VM numbers that > was the problem, or was it RMFPMS reporting Linux numbers?
As one of the people giving the answer Nick quoted...8-) The resource numbers that RMF-PM can collect from inside the Linux guest are only half the picture (and since they reflect what the virtual machine is doing, not the real machine), they are pretty much bogus. RMF-PM isn't completely useless -- it does report accurately about disk space utilization and swap utilization -- but it's view on processor consumption and other similiar usage is really warped. > How did folks > know that the numbers were inaccurate? They don't match the VM resource monitor stream. You need both real utilization AND the view from inside the virtual machine to get the whole picture. > Expensive to run: Was that in CPU > time? Any more info would be helpful! Yes. All the values have to be encoded into a XML representation, which is a very expensive process. The RMF-PM agent also is badly behaved about how and where it puts temp files and how it manages them. RMF-PM could be improved, but nobody's had the time. -- db ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
