On Tue, 22 Jul 2008, tl wrote:
> I search a simple variant to generate CPU,Load for MRTG.
> Actually I use following configuration.
> But I see CPU-usage of xen0 only, not a summary of all VM.
> - net-snmp does not support xen [1].
> - dtc-xen [2] show CPU-usage from each VM, but not summary.
> This view is confusing. [3]
> For monitorings from more than 3 physical machines unpractically.
> - enomalism [4] is too large for this small task.
>
> Someone has an idea or a finished (python|Perl)-script for this simple
task?
...meanwhile we created a solution:
http://www.fh-lausitz.de/launic/comp/xen/081202.xmstatmrtg-0.01
## xmstatmrtg, xm stats 4 mrtg
## - heiklehm noc_at_fh-lausitz.de
## - print cpu [%] (sum(xm list)/intervall/numcpu)
## - print load [%] /proc/loadavg
Torsten
_______________________________________________
Xen-users mailing list
Xen-users@lists.xensource.com
http://lists.xensource.com/xen-users
