Ater upgrade to 1.2 - poor KVM performance
- Thread starter nivlek
- Start date
What does "uptime" show?
When you start top, are there any processes taking excessive load?
Hello Mangoo,
the uptime looks to be a correct amount of time.
Processes seem to be using very little except KVM (Source ps -A)
Is there a better way to analyse this ? The install also took a long time. I would be prepared to reinstall if that would help.
Nivlek
"uptime" also shows system load, which is the most interesting thing it outputs
How many percent of CPU does kvm use in top?
Are you sure you are using kvm modules - could you also paste the output of lsmod?
Also, please paste the contents of the guest's configuration (in /etc/qemu-server).
"uptime" also shows system load, which is the most interesting thing it outputs
How many percent of CPU does kvm use in top?
Are you sure you are using kvm modules - could you also paste the output of lsmod?
Also, please paste the contents of the guest's configuration (in /etc/qemu-server).
Mangoo
I'm interested in learning more, but there are large gaps in my knowledge. I don't know how to obtain the answer re CPU % top. The other items are attached.
Nivlek
here is lsmod
proxmox:~# lsmod
Module Size Used by
kvm_intel 59496 1
kvm 203536 1 kvm_intel
vzethdev 23552 0
vznetdev 32904 0
simfs 14064 0
vzrst 158248 0
vzcpt 131512 0
tun 23168 2 vzrst,vzcpt
vzdquota 60016 0 [permanent]
vzmon 58008 4 vzethdev,vznetdev,vzrst,vzcpt
vzdev 12808 4 vzethdev,vznetdev,vzdquota,vzmon
xt_tcpudp 12160 0
xt_length 10752 0
ipt_ttl 10624 0
xt_tcpmss 11008 0
xt_TCPMSS 13440 0
iptable_mangle 13824 0
iptable_filter 13568 0
xt_multiport 12160 0
xt_limit 11904 0
ipt_tos 10368 0
ipt_REJECT 13824 0
ip_tables 33384 2 iptable_mangle,iptable_filter
x_tables 34056 10 xt_tcpudp,xt_length,ipt_ttl,xt_tcpmss,xt_TCPMSS,xt_multiport,xt_limit,ipt_tos,ipt_REJECT,ip_tables
ipv6 350720 37 vzrst,vzcpt,vzmon
bridge 75304 0
dm_snapshot 27616 0
dm_mirror 34688 0
snd_hda_intel 391592 0
snd_pcm 97288 1 snd_hda_intel
snd_timer 35464 1 snd_pcm
snd_page_alloc 19984 2 snd_hda_intel,snd_pcm
snd_hwdep 19848 1 snd_hda_intel
snd 79784 4 snd_hda_intel,snd_pcm,snd_timer,snd_hwdep
parport_pc 48168 0
parport 53388 1 parport_pc
psmouse 54684 0
soundcore 18208 1 snd
floppy 78056 0
dcdbas 18736 0
evdev 22912 0
thermal 26912 0
serio_raw 16388 0
tg3 135428 0
pcspkr 12160 0
button 18080 0
processor 49096 1 thermal
sg 49560 0
scsi_wait_scan 9984 0
virtio_blk 16264 0
virtio 14336 1 virtio_blk
dm_mod 80120 9 dm_snapshot,dm_mirror
usbhid 43616 0
hid 53312 1 usbhid
usb_storage 92480 0
libusual 30816 1 usb_storage
sd_mod 41088 3
sr_mod 27812 1
ide_disk 26624 0
ide_generic 9728 0 [permanent]
ide_cd 43552 0
cdrom 49064 2 sr_mod,ide_cd
ide_core 147608 3 ide_disk,ide_generic,ide_cd
uhci_hcd 38176 0
ehci_hcd 50188 0
usbcore 180784 6 usbhid,usb_storage,libusual,uhci_hcd,ehci_hcd
iTCO_wdt 22864 0
iTCO_vendor_support 12932 1 iTCO_wdt
ata_piix 31364 0
pata_acpi 17152 0
ata_generic 17156 0
ahci 40964 1
libata 189232 4 ata_piix,pata_acpi,ata_generic,ahci
i2c_i801 19356 0
i2c_core 36480 1 i2c_i801
shpchp 46236 0
pci_hotplug 43056 1 shpchp
mptsas 50192 0
mptscsih 50432 1 mptsas
mptbase 93284 2 mptsas,mptscsih
scsi_transport_sas 47744 1 mptsas
megaraid_sas 45756 2
scsi_mod 188856 10 sg,scsi_wait_scan,usb_storage,sd_mod,sr_mod,libata,mptsas,mptscsih,scsi_transport_sas,megaraid_sas
ohci1394 44212 0
ieee1394 114896 1 ohci1394
isofs 47784 0
msdos 19840 0
fat 68912 1 msdos
proxmox:~#
and here is the contents of the guest's configuration
ostype: wvista
memory: 4096
name: Vista_Business_64Bit
ide2: cdrom,media=cdrom
smp: 2
vlanu: rtl8139=62
4:47:A7:84:38bootdisk: scsi0
scsi0: vm-103-disk.qcow2
onboot: 0
Mangoo
I'm interested in learning more, but there are large gaps in my knowledge. I don't know how to obtain the answer re CPU % top. The other items are attached.
Nivlek
here is lsmod
proxmox:~# lsmod
Module Size Used by
kvm_intel 59496 1
kvm 203536 1 kvm_intel
proxmox:~#
and here is the contents of the guest's configuration
ostype: wvista
memory: 4096
name: Vista_Business_64Bit
ide2: cdrom,media=cdrom
smp: 2
vlanu: rtl8139=62
bootdisk: scsi0
scsi0: vm-103-disk.qcow2
onboot: 0
OK; so kvm module is loaded and used.
You may try using e1000 network card for your guest (you have to stop and start the guest after you edit that file).
top:
start it, press shift + P.
mangoo
I changed to e1000. Still slow.
top
Tasks: 117 total, 1 running, 116 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.9%us, 0.4%sy, 0.0%ni, 98.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 8170752k total, 6674044k used, 1496708k free, 46732k buffers
Swap: 7340024k total, 0k used, 7340024k free, 2194852k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
8681 root 20 0 4243m 3.9g 1600 S 5 50.4 0:50.46 kvm
8846 root 20 0 18828 1264 940 R 0 0.0 0:00.02 top
1 root 20 0 10316 748 620 S 0 0.0 0:04.84 init
2 root 15 -5 0 0 0 S 0 0.0 0:00.00 kthreadd
3 root RT -5 0 0 0 S 0 0.0 0:00.00 migration/0
4 root 15 -5 0 0 0 S 0 0.0 0:00.00 ksoftirqd/0
5 root RT -5 0 0 0 S 0 0.0 0:00.00 watchdog/0
6 root RT -5 0 0 0 S 0 0.0 0:00.00 migration/1
7 root 15 -5 0 0 0 S 0 0.0 0:00.00 ksoftirqd/1
8 root RT -5 0 0 0 S 0 0.0 0:00.00 watchdog/1
9 root RT -5 0 0 0 S 0 0.0 0:00.00 migration/2
10 root 15 -5 0 0 0 S 0 0.0 0:00.02 ksoftirqd/2
11 root RT -5 0 0 0 S 0 0.0 0:00.00 watchdog/2
12 root RT -5 0 0 0 S 0 0.0 0:00.00 migration/3
13 root 15 -5 0 0 0 S 0 0.0 0:00.00 ksoftirqd/3
14 root RT -5 0 0 0 S 0 0.0 0:00.00 watchdog/3
15 root RT -5 0 0 0 S 0 0.0 0:00.00 migration/4
16 root 15 -5 0 0 0 S 0 0.0 0:00.00 ksoftirqd/4
17 root RT -5 0 0 0 S 0 0.0 0:00.00 watchdog/4
18 root RT -5 0 0 0 S 0 0.0 0:00.00 migration/5
19 root 15 -5 0 0 0 S 0 0.0 0:00.00 ksoftirqd/5
20 root RT -5 0 0 0 S 0 0.0 0:00.00 watchdog/5
nivlek
M
mspock2010
Guest
Hi Mangoo,
Rather than allotting two CPUs to the guest, try to allot just one virtual cpu. However, this may require you to do a reinstallation of the guest (just a warning).
Good luck!
Rather than allotting two CPUs to the guest, try to allot just one virtual cpu. However, this may require you to do a reinstallation of the guest (just a warning).
Good luck!
I changed the number of CPUs from 2 to 1.
I don't believe that made any difference.
I am going to re-install the Guest
Re-installed with 2 CPUs defined - very slow as reported.
Re-installed with 1 CPU defined - very fast install and performance o.k.
Thanks for the tip.
So there appears to be a bug here ?
Nivlek
M
mspock2010
Guest
I believe that ALL virtualization engines recommend as few (if possible, just 1) virtual cpus as possible. I am not too familiar with the exact technical reason, but its like both physical CPUs have to be free to pass on that cycle to the guest as two CPUs.
Unlike when there is just one allotted, then either CPU can provided that particular cycle.
Glad that it helped you out.
Unlike when there is just one allotted, then either CPU can provided that particular cycle.
Glad that it helped you out.