Windows server 2012r2 unstable

dksys

New Member
Feb 4, 2015
11
0
1
Hi.
We have windows 2012 r2 Standart installed on proxmox
pveversion --verbose
proxmox-ve-2.6.32: 3.1-114 (running kernel: 2.6.32-26-pve)
pve-manager: 3.1-21 (running version: 3.1-21/93bf03d4)
pve-kernel-2.6.32-26-pve: 2.6.32-114
lvm2: 2.02.98-pve4
clvm: 2.02.98-pve4
corosync-pve: 1.4.5-1
openais-pve: 1.1.4-3
libqb0: 0.11.1-2
redhat-cluster-pve: 3.2.0-2
resource-agents-pve: 3.9.2-4
fence-agents-pve: 4.0.0-2
pve-cluster: 3.0-8
qemu-server: 3.1-8
pve-firmware: 1.0-23
libpve-common-perl: 3.0-8
libpve-access-control: 3.0-7
libpve-storage-perl: 3.0-17
pve-libspice-server1: 0.12.4-2
vncterm: 1.1-4
vzctl: 4.0-1pve4
vzprocps: 2.0.11-2
vzquota: 3.1-2
pve-qemu-kvm: 1.4-17
ksm-control-daemon: 1.1-1
glusterfs-client: 3.4.1-1

Host (128 GB RAM, CPU - intel E6-2650v2)

One VM running
boot: dcn
bootdisk: virtio0
cores: 7
cpu: host
ide2: local:iso/virtio-win-0.1-100.iso,media=cdrom,size=68314K
ide3: local:iso/DRP_14.6.iso,media=cdrom,size=1778352K
memory: 122880
name: w2012R2S-TERMINAL
net0: virtio=00:50:56:09:4E:62,bridge=vmbr1
net1: e1000=AE:55:60:F7:DE:02,bridge=vmbr0
onboot: 1
ostype: win8
sockets: 2
virtio0: LVMlocal:vm-201-disk-1,size=1000G

Windows server we use like terminal server (about 15 simultaneous rdp sessions).

At first day server works too bad (very high CPU load) but on second day everything was perfect, now - third day and server again works bad (and again too high CPU load - about 90-100%)

only one thing I found - too high Hardware Interrupts and DPCs

Any ideas will be helpful?
 
Is your VM memory dynamic or fixed? We had issues with both Windows server 2008/2012 R2 when memory was dynamic. After we fixed the amount of memory did not have any issue for last several months. On one Windows Serve 2012 R2 we have 22 RDP user and on the other Windows 2008 R2 about 14 RDP users.

It is advised that all Windows VM memory be fixed instead of dynamic.
 
Is your VM memory dynamic or fixed? We had issues with both Windows server 2008/2012 R2 when memory was dynamic. After we fixed the amount of memory did not have any issue for last several months. On one Windows Serve 2012 R2 we have 22 RDP user and on the other Windows 2008 R2 about 14 RDP users.

It is advised that all Windows VM memory be fixed instead of dynamic.


Thanks for fast reply, but I have fixed memory:(.
 
Hi. Thanks for help.
I upgraded proxmox:
proxmox-ve-2.6.32: 3.3-139 (running kernel: 2.6.32-34-pve)
pve-manager: 3.3-5 (running version: 3.3-5/bfebec03)
pve-kernel-2.6.32-34-pve: 2.6.32-140
pve-kernel-2.6.32-26-pve: 2.6.32-114
lvm2: 2.02.98-pve4
clvm: 2.02.98-pve4
corosync-pve: 1.4.7-1
openais-pve: 1.1.4-3
libqb0: 0.11.1-2
redhat-cluster-pve: 3.2.0-2
resource-agents-pve: 3.9.2-4
fence-agents-pve: 4.0.10-1
pve-cluster: 3.0-15
qemu-server: 3.3-3
pve-firmware: 1.1-3
libpve-common-perl: 3.0-19
libpve-access-control: 3.0-15
libpve-storage-perl: 3.0-25
pve-libspice-server1: 0.12.4-3
vncterm: 1.1-8
vzctl: 4.0-1pve6
vzprocps: 2.0.11-2
vzquota: 3.1-2
pve-qemu-kvm: 2.1-10
ksm-control-daemon: 1.1-1
glusterfs-client: 3.5.2-1

But no effect:(
Still too high interrupts, 3-4 terminal users loads CPU on 98%
interrupts.png
 
sorry and yes, its on the options tab. no poweroff/poweron needed.

ad2: no, just leave the default as this is a relative setting.
 
One more terrible day...
I disabled "use tablet for pointer", deleted all NICs with drivers, attached new NIC, installed virtio drivers for new NIC - still low performance.
Changed cache type to writeback - still low performance.
Please help)

 
can you test with pve-kernel 3.10 ?


Also, the pvetest repository have a new feature for cpu, "numa", which can help with big memory vm like you.
(need kernel 3.10 to works fine)

(I don't known if you are already in production or not)
 
Server is in production, therefore i can't use test repository

Just a shot in the dark. Have you looked over the power settings in the bios of the host. We have found that both HP/IBM restrict performance with power settings out of the box. Changing to max performance makes a significant difference. I don't think CPU units come into play with KVM but someone else might be able to better comment.
 
Just a shot in the dark. Have you looked over the power settings in the bios of the host. We have found that both HP/IBM restrict performance with power settings out of the box. Changing to max performance makes a significant difference. I don't think CPU units come into play with KVM but someone else might be able to better comment.

yes, server governor should be set to maxperformance.

cpuunits are not used if a virtio devices (disk,nic) exist in the guest vm.
 
Server is in production, therefore i can't use test repository

Do you have tested with less cpu and less ram in your guest ?

Do you have 1 x E6-2650v2 or 2x E6-2650v2 in your server ?


(if only 1 E6-2650v2 (8cores), 14 vcpus seem to be quite high, even if you enable hyperthreading)
 
proxmox-ve-2.6.32: 3.3-139 (running kernel: 2.6.32-34-pve)

You do not reboot after upgrade?

EDIT: sorry mistake mine is pve-kernel-2.6.32-37-pve but it seems to come from pvetest repository...
 
Last edited:
Do you have tested with less cpu and less ram in your guest ?

Do you have 1 x E6-2650v2 or 2x E6-2650v2 in your server ?


(if only 1 E6-2650v2 (8cores), 14 vcpus seem to be quite high, even if you enable hyperthreading)


Please in details. With hyperthreading i have 16 cores (8 cores 2 socets), i attached 7 cores 2 socets (14 vcpus) to VM, there is only one VM on this host - is it quite high?
 

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!

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!