Windows Server VM stürzt ab - KVM: entry failed, hardware error 0x80000021

T

tzzz90

New Member
May 10, 2022
14
1
3
Hallo,

folgendes Problem ist vorhanden, eine auf Proxmox migrierte Windows Server 2019 VM stürzt regelmäßig ab. In /var/log/syslog ist folgendes zu finden:

May 10 08:37:14 prox01 QEMU[1148]: KVM: entry failed, hardware error 0x80000021 May 10 08:37:14 prox01 QEMU[1148]: If you're running a guest on an Intel machine without unrestricted mode May 10 08:37:14 prox01 kernel: [87072.394575] set kvm_intel.dump_invalid_vmcs=1 to dump internal KVM state. May 10 08:37:14 prox01 QEMU[1148]: support, the failure can be most likely due to the guest entering an invalid May 10 08:37:14 prox01 QEMU[1148]: state for Intel VT. For example, the guest maybe running in big real mode May 10 08:37:14 prox01 QEMU[1148]: which is not supported on less recent Intel processors. May 10 08:37:14 prox01 QEMU[1148]: EAX=002b0c36 EBX=1cfd9180 ECX=00000000 EDX=00000000 May 10 08:37:14 prox01 QEMU[1148]: ESI=1cfe9200 EDI=33c19080 EBP=00000000 ESP=23db2660 May 10 08:37:14 prox01 QEMU[1148]: EIP=00008000 EFL=00000002 [-------] CPL=0 II=0 A20=1 SMM=1 HLT=0 May 10 08:37:14 prox01 QEMU[1148]: ES =0000 00000000 ffffffff 00809300 May 10 08:37:14 prox01 QEMU[1148]: CS =c200 7ffc2000 ffffffff 00809300 May 10 08:37:14 prox01 QEMU[1148]: SS =0000 00000000 ffffffff 00809300 May 10 08:37:14 prox01 QEMU[1148]: DS =0000 00000000 ffffffff 00809300 May 10 08:37:14 prox01 QEMU[1148]: FS =0000 00000000 ffffffff 00809300 May 10 08:37:14 prox01 QEMU[1148]: GS =0000 00000000 ffffffff 00809300 May 10 08:37:14 prox01 QEMU[1148]: LDT=0000 00000000 000fffff 00000000 May 10 08:37:14 prox01 QEMU[1148]: TR =0040 1cfec000 00000067 00008b00 May 10 08:37:14 prox01 QEMU[1148]: GDT= 1cfedfb0 00000057 May 10 08:37:14 prox01 QEMU[1148]: IDT= 00000000 00000000 May 10 08:37:14 prox01 QEMU[1148]: CR0=00050032 CR2=ba0414d4 CR3=b9036002 CR4=00000000 May 10 08:37:14 prox01 QEMU[1148]: DR0=0000000000000000 DR1=0000000000000000 DR2=0000000000000000 DR3=0000000000000000 May 10 08:37:14 prox01 QEMU[1148]: DR6=00000000fffe0ff0 DR7=0000000000000400 May 10 08:37:14 prox01 QEMU[1148]: EFER=0000000000000000 May 10 08:37:14 prox01 QEMU[1148]: Code=kvm: ../hw/core/cpu-sysemu.c:77: cpu_asidx_from_attrs: Assertion `ret < cpu->num_ases && ret >= 0' failed.

pveversion

Code: 
pve-manager/7.2-3/c743d6c1 (running kernel: 5.15.35-1-pve)

qm config

Code: 
agent: 0
balloon: 0
bios: ovmf
boot:
cores: 8
cpu: host
efidisk0: local-lvm:vm-101-disk-1,efitype=4m,pre-enrolled-keys=1,size=4M
machine: pc-q35-5.0
memory: 16000
meta: creation-qemu=6.2.0,ctime=1651869005
name: Name
net0: e1000=5A:4E:0B:08:47:D9,bridge=vmbr0
numa: 0
onboot: 1
ostype: win10
scsi0: local-lvm:vm-101-disk-0,size=80G
scsi1: local-lvm:vm-101-disk-2,size=500G
scsihw: virtio-scsi-pci
smbios1: uuid=d5161f55-533b-41e3-ab1a-2b06ebd24a4a
sockets: 1
vmgenid: da29ce18-4168-4aca-a3c6-2b8b7254d88b

Finde dazu nicht wirklich eine Lösung, der Maschinentyp q35-5.0 daher, weil sonst keine Netzwerkkarte vorhanden ist.

CPU

Code: 
description: CPU
          product: Intel(R) Xeon(R) E-2244G CPU @ 3.80GHz
          vendor: Intel Corp.
          physical id: 57
          bus info: cpu@0
          version: Intel(R) Xeon(R) E-2244G CPU @ 3.80GHz
          serial: To Be Filled By O.E.M.
          slot: CPU1
          size: 3231MHz
          capacity: 4005MHz
          width: 64 bits
          clock: 100MHz

Vielen Dank im Vorraus!

VG
tzzz
 
Hallo zusammen,

seit dem Update auf die 7.2.3 habe ich plötzlich auch sporadisch auf migrierten 2016ern Servern (von Hyper-V migriert) bei 2 Kunden.
wurde das schon untersucht?

schönen Gruß
 
Guten Morgen, leider hat das nicht geholfen, die Maschine stürzt weiterhin regelmäßig ab.


VG
 
Hallo zusammen,
irgendwas haben die im Kernel geändert oder im KVM und betrifft nur Intel Xeon neuerer Generationen.
Wenn hier keiner weiter hilft muss man wohl ein Ticket aufmachen?
hier mal ein Auszug eines Crashes heut morgen von einem Domaincontroller

Code: 
May 12 08:21:11 pve kernel: device tap2012013i0 entered promiscuous mode
May 12 08:21:11 pve kernel: vmbr1: port 2(tap2012013i0) entered blocking state
May 12 08:21:11 pve kernel: vmbr1: port 2(tap2012013i0) entered disabled state
May 12 08:21:11 pve kernel: vmbr1: port 2(tap2012013i0) entered blocking state
May 12 08:21:11 pve kernel: vmbr1: port 2(tap2012013i0) entered forwarding state
May 12 08:21:11 pve pvedaemon[1871989]: <root@pam> end task UPID:pve:001C9088:02C4BE0E:627CA757:qmstart:2012013:root@pam: OK
May 12 08:21:15 pve pvedaemon[1516922]: VM 2012013 qmp command failed - VM 2012013 qmp command 'guest-ping' failed - got timeout
May 12 08:21:34 pve QEMU[1872029]: KVM: entry failed, hardware error 0x80000021
May 12 08:21:34 pve QEMU[1872029]: If you're running a guest on an Intel machine without unrestricted mode
May 12 08:21:34 pve QEMU[1872029]: support, the failure can be most likely due to the guest entering an invalid
May 12 08:21:34 pve QEMU[1872029]: state for Intel VT. For example, the guest maybe running in big real mode
May 12 08:21:34 pve QEMU[1872029]: which is not supported on less recent Intel processors.
May 12 08:21:34 pve QEMU[1872029]: EAX=0008c1b8 EBX=f5158180 ECX=00000000 EDX=00000000
May 12 08:21:34 pve QEMU[1872029]: ESI=f51bfa40 EDI=ad2ac080 EBP=00000000 ESP=38ee2660
May 12 08:21:34 pve QEMU[1872029]: EIP=00008000 EFL=00000002 [-------] CPL=0 II=0 A20=1 SMM=1 HLT=0
May 12 08:21:34 pve QEMU[1872029]: ES =0000 00000000 ffffffff 00809300
May 12 08:21:34 pve QEMU[1872029]: CS =be00 7ffbe000 ffffffff 00809300
May 12 08:21:34 pve QEMU[1872029]: SS =0000 00000000 ffffffff 00809300
May 12 08:21:34 pve QEMU[1872029]: DS =0000 00000000 ffffffff 00809300
May 12 08:21:34 pve QEMU[1872029]: FS =0000 00000000 ffffffff 00809300
May 12 08:21:34 pve QEMU[1872029]: GS =0000 00000000 ffffffff 00809300
May 12 08:21:34 pve QEMU[1872029]: LDT=0000 00000000 000fffff 00000000
May 12 08:21:34 pve QEMU[1872029]: TR =0040 f638b000 00000067 00008b00
May 12 08:21:34 pve QEMU[1872029]: GDT=     f638c000 0000007f
May 12 08:21:34 pve QEMU[1872029]: IDT=     00000000 00000000
May 12 08:21:34 pve QEMU[1872029]: CR0=00050032 CR2=b66e6ff0 CR3=001ab000 CR4=00000000
May 12 08:21:34 pve QEMU[1872029]: DR0=0000000000000000 DR1=0000000000000000 DR2=0000000000000000 DR3=0000000000000000
May 12 08:21:34 pve QEMU[1872029]: DR6=00000000ffff0ff0 DR7=0000000000000400
May 12 08:21:34 pve QEMU[1872029]: EFER=0000000000000000
May 12 08:21:34 pve QEMU[1872029]: Code=kvm: ../hw/core/cpu-sysemu.c:77: cpu_asidx_from_attrs: Assertion `ret < cpu->num_ases && ret >= 0' failed.
May 12 08:21:34 pve kernel: set kvm_intel.dump_invalid_vmcs=1 to dump internal KVM state.
May 12 08:21:34 pve kernel: vmbr1: port 2(tap2012013i0) entered disabled state
May 12 08:21:34 pve kernel: vmbr1: port 2(tap2012013i0) entered disabled state
May 12 08:21:34 pve systemd[1]: 2012013.scope: Succeeded.
May 12 08:21:34 pve systemd[1]: 2012013.scope: Consumed 23.347s CPU time.
May 12 08:21:34 pve qmeventd[1872178]: Starting cleanup for 2012013
May 12 08:21:34 pve qmeventd[1872178]: Finished cleanup for 2012013
May 12 08:22:03 pve pvedaemon[1872370]: start VM 2012013: UPID:pve:001C91F2:02C4D25B:627CA78A:qmstart:2012013:root@pam:
May 12 08:22:03 pve pvedaemon[1738180]: <root@pam> starting task UPID:pve:001C91F2:02C4D25B:627CA78A:qmstart:2012013:root@pam:
May 12 08:22:03 pve systemd[1]: Started 2012013.scope.

Seit dem läuft die VM wieder, nur wie lange?
Gruß
 
Hallo,

hier heute auch wieder das Problem und in der englischsprachigen Sektion scheint das Problem wohl auch öfters aufzutreten.

VG
 
Nicht nur bei den neuen Xeon, hier auch mit einem neuen i5-12500 (alder lake).

Bei uns führt das im Übrigen dazu, dass nur noch Fragmente der Proxmox-GUI im Browser angezeigt werden und insbesondere die Graphen ab dem Zeitpunkt der Fehlermeldung gar nicht mehr dargestellt werden. Ein Zugriff über SSH ist aber noch möglich, es schmiert auch nur eine VM ab, dies allerdings regelmäßig.

Wir haben nahezu die selben log-Fehlermeldungen wie fmtech:

Code: 
May 29 03:00:19 EX43 QEMU[1790514]: KVM: entry failed, hardware error 0x80000021
May 29 03:00:19 EX43 QEMU[1790514]: If you're running a guest on an Intel machine without unrestricted mode
May 29 03:00:19 EX43 QEMU[1790514]: support, the failure can be most likely due to the guest entering an invalid
May 29 03:00:19 EX43 QEMU[1790514]: state for Intel VT. For example, the guest maybe running in big real mode
May 29 03:00:19 EX43 QEMU[1790514]: which is not supported on less recent Intel processors.
May 29 03:00:19 EX43 QEMU[1790514]: EAX=0019acb6 EBX=1101b429 ECX=00331d7a EDX=00000000
May 29 03:00:19 EX43 QEMU[1790514]: ESI=53677f80 EDI=0019effc EBP=595af540 ESP=57297fb0
May 29 03:00:19 EX43 QEMU[1790514]: EIP=00008000 EFL=00000002 [-------] CPL=0 II=0 A20=1 SMM=1 HLT=0
May 29 03:00:19 EX43 QEMU[1790514]: ES =0000 00000000 ffffffff 00809300
May 29 03:00:19 EX43 QEMU[1790514]: CS =ba00 7ffba000 ffffffff 00809300
May 29 03:00:19 EX43 kernel: set kvm_intel.dump_invalid_vmcs=1 to dump internal KVM state.
May 29 03:00:19 EX43 QEMU[1790514]: SS =0000 00000000 ffffffff 00809300
May 29 03:00:19 EX43 QEMU[1790514]: DS =0000 00000000 ffffffff 00809300
May 29 03:00:19 EX43 QEMU[1790514]: FS =0000 00000000 ffffffff 00809300
May 29 03:00:19 EX43 QEMU[1790514]: GS =0000 00000000 ffffffff 00809300
May 29 03:00:19 EX43 QEMU[1790514]: LDT=0000 00000000 000fffff 00000000
May 29 03:00:19 EX43 QEMU[1790514]: TR =0040 5727c000 00000067 00008b00
May 29 03:00:19 EX43 QEMU[1790514]: GDT=     5727dfb0 00000057
May 29 03:00:19 EX43 QEMU[1790514]: IDT=     00000000 00000000
May 29 03:00:19 EX43 QEMU[1790514]: CR0=00050032 CR2=00000000 CR3=e7867000 CR4=00000000
May 29 03:00:19 EX43 QEMU[1790514]: DR0=0000000000000000 DR1=0000000000000000 DR2=0000000000000000 DR3=0000000000000000
May 29 03:00:19 EX43 QEMU[1790514]: DR6=00000000ffff4ff0 DR7=0000000000000400
May 29 03:00:19 EX43 QEMU[1790514]: EFER=0000000000000000
May 29 03:00:19 EX43 QEMU[1790514]: Code=kvm: ../hw/core/cpu-sysemu.c:77: cpu_asidx_from_attrs: Assertion `ret < cpu->num_ases && ret >= 0' failed.
 
You must log in or register to reply here.

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Quick Navigation

Home Get Subscription Wiki Downloads Proxmox Customer Portal About

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!
Community platform by XenForo® © 2010-2024 XenForo Ltd.
Top Bottom
Back