virtualbox.org

Hello, I use I5-3210 Win7, 16 Ram, 1 GB SSD, 1 GB HDD, 4 core cpu. I have Win7, Win8, Win10 virtual machine. Last day all machine is working good. But now not working. Windows is open and cpu usage is over %50. I can not use virtual machine. After machine is not open and I see on screen message "send image file log to community group"
What can I do this problem ? (picture and tree log file in rar files)

You had a Guru Meditation:

00:00:20.773611 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
00:00:20.773614 !!
00:00:20.773614 !!         VCPU1: Guru Meditation 1155 (VINF_EM_TRIPLE_FAULT)
00:00:20.773640 !!
00:00:20.773646 !! Skipping ring-0 registers and stack, rcErr=VINF_EM_TRIPLE_FAULT

Triple fault means the guest tried something so unusual it blew past all the error checking Virtualbox has.

The message right before the Guru spoke is:

00:00:20.187669 VMMDev: vmmDevHeartbeatFlatlinedTimer: Guest seems to be unresponsive. Last heartbeat received 7 seconds ago
00:00:20.189958 PGMR3PhysAllocateLargePage: allocating large pages takes too long (last attempt 6525 ms; nr of timeouts 1); DISABLE

This part: "PhysAllocateLargePage: allocating large pages" makes me wonder if, even though you have enough free memory:

00:00:01.342036 Host RAM: 16246MB (15.8GB) total, 13823MB (13.4GB) available
00:00:01.510905   RamSize           <integer> = 0x0000000200000000 (8 589 934 592, 8 192 MB, 8.0 GB)

you might not have it all in one big 8-GB-wide chunk. I might be wrong, there's others on the forum who'll pull out a bunch of hex from the log and figure exactly what's up. But try rebooting your host and then running the guest first thing after logging in.

One other thing, your 4-core I5 is really a two-core hyperthreaded CPU:

00:00:01.849500 CPUM: Physical host cores: 2
00:00:01.849825 Full Name:   "Intel(R) Core(TM) i5-3210M CPU @ 2.50GHz"

See the link to Intel's website, the line "# of cores". Hyperthreading allows extra threads through at the same time if the threads are doing different things and the core can allow it. Virtualbox does not benefit from hyperthreading, so you have only two cores for guest purposes.

Technically you have both cores in the guest:

00:00:01.510900   NumCPUs <integer> = 0x0000000000000002 (2)

Whichmeans that if the guest goes full-throttle on both cores, the host could suffer or crash. That said, I have the same setup in my laptop, two-core hyperthreaded i5, (to me that's really an i3 with marketing paint on it, my other PC is a real I5 ). I run two-core guests on it just fine. I am aware of the maxed-out guest core count and am ready to know why my laptop will get sluggish one day if it ever does. It's a risk assessment.

So you don't have to change anything, just be aware what might happen one day, and know that you have no more room for another guest if you're running this one at its present core count.

scottgus1 wrote:there's others on the forum who'll pull out a bunch of hex from the log and figure exactly what's up.

Did someone call my name?

Sorry to have to disappoint you, but this time the error state is not so obvious to me. What I can tell from the logs:

The two crashes happen at different relative times soon after booting from the hard disk begins. vCPU 0 is in 64 bit mode and has been running for several seconds. vCPU 1 is (still?) in 32 bit mode, has been running for at most 20 microseconds and crashes. (The running times can be observed in the log lines containing /TM/CPU/.) A hint to the type of error is

VBox.log wrote:00:00:20.781422 CPUM1: Disas -> VERR_PAGE_TABLE_NOT_PRESENT

My educated guess is that the error occurs when vCPU 1 is put into operation. Someone could ask a VirtualBox core developer for further directions ...