VM windows shutdown by itself

[Thatoo]

Hello,

Two days ago, I found a VM windows off without reason. I first thought it had switch off itself after some obscure updates but apparently no.
Then I look at Proxmox's logs and I found the following line I can't understand :

Sep 21 10:53:13 esfr smartd[2999314]: Device: /dev/sdb [SAT], SMART Usage Attribute: 194 Temperature_Celsius changed from 73 to 74
Sep 21 10:53:13 esfr smartd[2999314]: Device: /dev/sdd [SAT], SMART Usage Attribute: 194 Temperature_Celsius changed from 64 to 63
Sep 21 10:56:44 esfr QEMU[2593039]: KVM: entry failed, hardware error 0x80000021
Sep 21 10:56:45 esfr QEMU[2593039]: If you're running a guest on an Intel machine without unrestricted mode
Sep 21 10:56:45 esfr QEMU[2593039]: support, the failure can be most likely due to the guest entering an invalid
Sep 21 10:56:45 esfr QEMU[2593039]: state for Intel VT. For example, the guest maybe running in big real mode
Sep 21 10:56:45 esfr QEMU[2593039]: which is not supported on less recent Intel processors.
Sep 21 10:56:45 esfr QEMU[2593039]: EAX=00000000 EBX=81bd06d4 ECX=00000000 EDX=7da80a70
Sep 21 10:56:45 esfr QEMU[2593039]: ESI=5a6f7ff0 EDI=0005f81d EBP=9ec63b60 ESP=5f1f4fb0
Sep 21 10:56:45 esfr QEMU[2593039]: EIP=00008000 EFL=00000002 [-------] CPL=0 II=0 A20=1 SMM=1 HLT=0
Sep 21 10:56:45 esfr kernel: [3437592.021910] set kvm_intel.dump_invalid_vmcs=1 to dump internal KVM state.
Sep 21 10:56:45 esfr QEMU[2593039]: ES =0000 00000000 ffffffff 00809300
Sep 21 10:56:45 esfr QEMU[2593039]: CS =9e00 7ff9e000 ffffffff 00809300
Sep 21 10:56:45 esfr QEMU[2593039]: SS =0000 00000000 ffffffff 00809300
Sep 21 10:56:45 esfr QEMU[2593039]: DS =0000 00000000 ffffffff 00809300
Sep 21 10:56:45 esfr QEMU[2593039]: FS =0000 00000000 ffffffff 00809300
Sep 21 10:56:45 esfr QEMU[2593039]: GS =0000 00000000 ffffffff 00809300
Sep 21 10:56:45 esfr QEMU[2593039]: LDT=0000 00000000 000fffff 00000000
Sep 21 10:56:45 esfr QEMU[2593039]: TR =0040 5f1db000 00000067 00008b00
Sep 21 10:56:45 esfr QEMU[2593039]: GDT= 5f1dcfb0 00000057
Sep 21 10:56:45 esfr QEMU[2593039]: IDT= 00000000 00000000
Sep 21 10:56:45 esfr QEMU[2593039]: CR0=00050032 CR2=1d26d25a CR3=7afe0000 CR4=00000000
Sep 21 10:56:45 esfr QEMU[2593039]: DR0=0000000000000000 DR1=0000000000000000 DR2=0000000000000000 DR3=0000000000000000
Sep 21 10:56:45 esfr QEMU[2593039]: DR6=00000000ffff0ff1 DR7=0000000000000400
Sep 21 10:56:45 esfr QEMU[2593039]: EFER=0000000000000000
Sep 21 10:56:45 esfr QEMU[2593039]: Code=kvm: ../hw/core/cpu-sysemu.c:77: cpu_asidx_from_attrs: Assertion `ret < cpu->num_ases && ret >= 0' failed.
Sep 21 10:56:45 esfr kernel: [3437592.630039] fwbr201i0: port 2(tap201i0) entered disabled state
Sep 21 10:56:45 esfr kernel: [3437592.630563] fwbr201i0: port 2(tap201i0) entered disabled state
Sep 21 10:56:45 esfr qmeventd[793755]: Starting cleanup for 201
Sep 21 10:56:45 esfr kernel: [3437592.923036] fwbr201i0: port 1(fwln201i0) entered disabled state
Sep 21 10:56:45 esfr kernel: [3437592.923387] vmbr0: port 2(fwpr201p0) entered disabled state
Sep 21 10:56:45 esfr kernel: [3437592.923910] device fwln201i0 left promiscuous mode
Sep 21 10:56:45 esfr kernel: [3437592.924232] fwbr201i0: port 1(fwln201i0) entered disabled state
Sep 21 10:56:46 esfr kernel: [3437593.145555] device fwpr201p0 left promiscuous mode
Sep 21 10:56:46 esfr kernel: [3437593.145894] vmbr0: port 2(fwpr201p0) entered disabled state
Sep 21 10:56:46 esfr qmeventd[793755]: Finished cleanup for 201
Sep 21 10:56:47 esfr systemd[1]: 201.scope: Succeeded.
Sep 21 10:56:47 esfr systemd[1]: 201.scope: Consumed 6d 7h 50min 707ms CPU time.

I guess, it is the lines

Sep 21 10:56:45 esfr qmeventd[793755]: Starting cleanup for 201

and

Sep 21 10:56:46 esfr qmeventd[793755]: Finished cleanup for 201
Sep 21 10:56:47 esfr systemd[1]: 201.scope: Succeeded.
Sep 21 10:56:47 esfr systemd[1]: 201.scope: Consumed 6d 7h 50min 707ms CPU time.

that confirm that it is Proxmox that switch off the VM but I'd like to know why? and how to prevent it?

 

[Thatoo]

I found someone who seem to have a similar issue : https://forum.proxmox.com/threads/qmeventd-shutdown-vm.92279

 

[Thatoo]

It could be because the QEMU Guest Agent was not running automatically on the VM. I changed that in the Services of windows.

 

[Thatoo]

and the problem would be more linked to this issue apparently : https://forum.proxmox.com/threads/vm-shutdown-kvm-entry-failed-hardware-error-0x80000021.109410/