Show CPU/memory usage in Qube manager again

[CodeFetch]

Back in 2018 the CPU and memory usage columns were removed from the Qube manager.
Since then it annoys me I only see these values in the widget. I see it as a regression as one could select which columns to display in the "View" tab of Qube manager. Please reintroduce these.

[andrewdavidwong]

This was already the initial request in #4531, but I don't know whether @marmarta and/or other devs decided that having these values only in the Domains widget was still what they wanted or whether the OP of #4531 just didn't realize they were already in the widget and was satisfied by that once it was pointed out to them.

I'll defer to @marmarta on whether to keep this open.

[marmarta]

They were removed due to the complexity of the code that handled them (during the general qube manager rewrite; it was a terrifying piece of jungle-style code) and the fact that they were available in the widget. Generally, I'm not sure if this is the best way to use our resources, but if someone wants to do this, feel free to do it.

[marmarek]

For anybody wanting to work on this:

• how to do graphs can be found in git history of the qubes-manager - it was there before R4.0 release, so looking at release3.2 branch might be good enough
• how to get the data for the graphs - look at the domains widget that already gets that info

[alimirjamali]

Question on admin.vm.Stats API call. The online documentation mentions:

Method admin.vm.Stats returns vm-stats events every stats_interval seconds, for every running VM. Parameters of vm-stats events:

memory_kb - memory usage in kB
cpu_time - absolute CPU time (in milliseconds) spent by the VM since its startup, normalized for one CPU
cpu_usage - CPU usage in percents

While my observation of running:

qubesd-query dom0 admin.vm.Stats dom0 --empty

Shows that CPU usage percentage (cpu_usage) is a rounded integer (either 0 or 1). Is the API broken? (We also have the undocumented cpu_usage_raw)

P.S.: Nevermind. It is working. Just rounded (CPU usage was really low on time of testing, I thought it to be core count).