Historical CPU load, etc

I know that collectinfo.pl collects historical CPU load and other things… where do I view that at in Virtualmin GPL?

The graphs don’t exist in Virtualmin GPL. But, as you’ve noticed, we’ve begun to move the collection out of being exclusive to Virtualmin Professional (it’s actually moving into Webmin, by and large). We have vague plans for making some of those currently Pro-only features available to as many folks as possible…including Virtualmin GPL and Webmin users. We haven’t quite figured out the specifics of it yet, and the code isn’t written yet, though. :wink:

One of the reasons for this is that Webmin is now our “agent” for Cloudmin. Cloudmin doesn’t need to run on all systems in a cloud (just the administrative master), but Webmin can easily do so, and for free. So, for now, the data is being gathered for the benefit of Cloudmin. If you aren’t using Cloudmin, you’ll get no graphs.

The way it is shaping up however, I suspect that in a couple of months, you’ll have many ways to work with that data, including graphs (a big part of our todo list is making analytics data for large numbers of physical machines and virtual machines useful and easy to browse). Some of those will be in the Open Source projects, while others will be standalone commercial products.

I’d be curious what folks think of some of the following ideas for such tools:

iPhone and Android status and monitoring application with alerts (this would be commercial due to the nature of that market)

RSS and other “realtime” protocol monitoring options…add status monitors and data points to your RSS reader. This one would be Open Source, and probably become a standard part of Webmin. It’s a reasonably low-level of effort, since it doesn’t require graph library integration and such.

A dedicated system data analytics module that works with this data, as well as data from other sources. This one would probably need to be commercial, as it is a major development project.

Which of these three options seems most awesome to you folks? Any other ideas for making the data we’re gathering more generally useful?