Background:
In the past when my guest VM froze, I looked at windows task manager and saw very little ram. This led me to conclude "the host put the VM out of physical memory and into virtual memory (i.e. paging) --> therefore the VM froze because it was now running out of paging instead of physical memory".
Current:
Most recently my guest VM froze and looking at task manager revealed that there was plenty of "available memory" 1.2 GB. I have attached the screenshot of my host windows machine and the guest Fedora 14 VM.
My observations:
- the host has plenty of available memory (1.2 GB)
- the guest has plenty of available memory
- the screenshot of the guest... is only available because it was by complete chance on the top and visible
- the host CPU is pegged for 2/4 cores
- guest CPU is not pegged at all
I closed the VM and my host's task manager showed:
- physical memory available went up by exactly the same amount of memory allocated to the guest VM
- paging went went down by exactly the same amount of memory allocated to the guest VM
- CPU usage went from super high to almost nothing
Given the description, why do you think the guest VM froze? I just can not figure this out and I don't have any debugging tools to see what the problem is!
- Is there someway to show that the "virtualbox" process is heavily using the hardisk (i.e. the process is in paging now instead of physical memory?
- How would I show virtual-box is "stuck in paging"?
- (I have heard of perfmon... but could not figure it out. so please elaborate more than "use perfmon" and say what kind of counters to look at)