Windows XP KVM limited at 100mbps

S

shadowq

Renowned Member
Mar 12, 2013
32
2
73
Hello,

I've been using proxmox for a while with several OpenVZ containers, but needed to use a Windows guest. So have installed windows XP as a guest, followed http://pve.proxmox.com/wiki/Windows_XP_Guest_Notes to change the network and disk to virtio, and have made registry changes at http://www.linux-kvm.org/page/WindowsGuestDrivers/kvmnet/registry. The Windows guest indicates that it's network connection is 1Gbps.

However- iperf results show that it seems to be capping at 100mbps to any other guest or to the host. The OpenVZ guests are maxing out at 1Gbps per thread (and ~5Gbps with multiple threads). I haven't got any other KVM guests running to test them, but could do that if it's not something obvious.

Code: 
root@s1:~# pveperf
CPU BOGOMIPS:      39996.06
REGEX/SECOND:      918234
HD SIZE:           94.49 GB (/dev/mapper/pve-root)
BUFFERED READS:    136.25 MB/sec
AVERAGE SEEK TIME: 11.75 ms
FSYNCS/SECOND:     608.10
DNS EXT:           117.80 ms
DNS INT:           37.66 ms (kjhosting.com)

Code: 
root@s1:~# pveversion --verbose
pve-manager: 2.1-1 (pve-manager/2.1/f9b0f63a)
running kernel: 2.6.32-11-pve
proxmox-ve-2.6.32: 2.0-66
pve-kernel-2.6.32-11-pve: 2.6.32-66
lvm2: 2.02.95-1pve2
clvm: 2.02.95-1pve2
corosync-pve: 1.4.3-1
openais-pve: 1.1.4-2
libqb: 0.10.1-2
redhat-cluster-pve: 3.1.8-3
resource-agents-pve: 3.9.2-3
fence-agents-pve: 3.1.7-2
pve-cluster: 1.0-26
qemu-server: 2.0-39
pve-firmware: 1.0-16
libpve-common-perl: 1.0-27
libpve-access-control: 1.0-21
libpve-storage-perl: 2.0-18
vncterm: 1.0-2
vzctl: 3.0.30-2pve5
vzprocps: 2.0.11-2
vzquota: 3.0.12-3
pve-qemu-kvm: 1.0-9
ksm-control-daemon: 1.1-1

Any ideas / suggestions would be muchly appreciated!

Cheers,
Jarrod.
 
Thanks for the reply Tom!

I have looked at that, and strongly considered updating. However, I was worried that there may be problems (and it's a production server).

Has there been many reported problems in upgrading?

Thanks again,
Jarrod.
 
the general recommendation is always using the latest stable version.

if the upgrade has issues to your box, how should we know it? test it!

=> make sure you have a valid backup of your data (this is a general recommendation) and do the upgrade as described.
 
I can second to that - very strange, indeed. Windows XP SP3 guest is on latest virtio drivers from from virtio-win-0.1-52.iso (from fedoraproject.org), offloads are disabled and yet outgoing traffic can not reach above 110-120 Mbit/s while incoming traffic reaches aroung 450-500 Mbit/s (measured with iperf) - not perfect but I can live with that. Traffic measurement on the host gives me figures around 930-950 Mbit/s.

My suspicion is some regression in the netkvm driver, however did not found too much information about it - found this on public Redhat bugzilla https://bugzilla.redhat.com/show_bug.cgi?id=884425 however they talk about RX performance regression.

Tried to downgrade netkvm drivers taking developers build from http://people.redhat.com/vrozenfe/ - did not help very much.

Host info:
Code: 
root@v1:~# pveperf 
CPU BOGOMIPS:      127678.40
REGEX/SECOND:      1035693
HD SIZE:           19.69 GB (/dev/mapper/pve-root)
BUFFERED READS:    338.72 MB/sec
AVERAGE SEEK TIME: 5.84 ms
FSYNCS/SECOND:     211.96
DNS EXT:           38.16 ms
DNS INT:           1.20 ms (domain.com]

Code: 
root@v1:~# pveversion --verbose
pve-manager: 2.3-13 (pve-manager/2.3/7946f1f1)
running kernel: 2.6.32-18-pve
proxmox-ve-2.6.32: 2.3-88
pve-kernel-2.6.32-16-pve: 2.6.32-82
pve-kernel-2.6.32-18-pve: 2.6.32-88
lvm2: 2.02.95-1pve2
clvm: 2.02.95-1pve2
corosync-pve: 1.4.4-4
openais-pve: 1.1.4-2
libqb: 0.10.1-2
redhat-cluster-pve: 3.1.93-2
resource-agents-pve: 3.9.2-3
fence-agents-pve: 3.1.9-1
pve-cluster: 1.0-36
qemu-server: 2.3-17
pve-firmware: 1.0-21
libpve-common-perl: 1.0-48
libpve-access-control: 1.0-26
libpve-storage-perl: 2.3-6
vncterm: 1.0-3
vzctl: 4.0-1pve2
vzprocps: 2.0.11-2
vzquota: 3.1-1
pve-qemu-kvm: 1.4-6
ksm-control-daemon: 1.1-1
 
tom said:
the general recommendation is always using the latest stable version.

if the upgrade has issues to your box, how should we know it? test it!

=> make sure you have a valid backup of your data (this is a general recommendation) and do the upgrade as described.
Click to expand...

Thanks again for your reply Tom! However, I wasn't asking if it will have problems or issues on our server- I was asking if you have had any issues reported by other users upgrading?

This project isn't as urgent anymore- as we've decided to use an EC2 instance closer to the client due to latency.

Cheers,
Jarrod.
 
You must log in or register to reply here.
Top Bottom
Back