Wrong CPU-usage

[gosha]

Hi!

VM's CPU-usage (in last GUI) is not displayed correctly:



--
Best regards,
Gosha

 

[fabian]

Please post the complete output of

• pveversion -v
• qm config ID

 

[gosha]

Ok.

Please post the complete output of

• pveversion -v
• qm config ID

Ok:

Spoiler: pveversion -v

pveversion -v
proxmox-ve: 4.1-48 (running kernel: 4.4.6-1-pve)
pve-manager: 4.1-34 (running version: 4.1-34/8887b0fd)
pve-kernel-4.4.6-1-pve: 4.4.6-48
pve-kernel-4.2.6-1-pve: 4.2.6-36
pve-kernel-4.2.8-1-pve: 4.2.8-41
pve-kernel-4.2.3-2-pve: 4.2.3-22
lvm2: 2.02.116-pve2
corosync-pve: 2.3.5-2
libqb0: 1.0-1
pve-cluster: 4.0-39
qemu-server: 4.0-72
pve-firmware: 1.1-8
libpve-common-perl: 4.0-59
libpve-access-control: 4.0-16
libpve-storage-perl: 4.0-50
pve-libspice-server1: 0.12.5-2
vncterm: 1.2-1
pve-qemu-kvm: 2.5-13
pve-container: 1.0-61
pve-firewall: 2.0-25
pve-ha-manager: 1.0-28
ksm-control-daemon: 1.2-1
glusterfs-client: 3.5.2-2+deb8u1
lxc-pve: 1.1.5-7
lxcfs: 2.0.0-pve2
cgmanager: 0.39-pve1
criu: 1.6.0-1
fence-agents-pve: 4.0.20-1

Spoiler: qm config 105

balloon: 8192
boot: c
bootdisk: virtio0
cores: 6
cpu: SandyBridge
memory: 16384
name: ts1
net0: virtio=4E:5D:97:43:13:09,bridge=vmbr0
net1: virtio=C6:E37:6A:B3:74,bridge=vmbr2
numa: 0
ostype: win7
smbios1: uuid=a2a5f12e-51b5-4431-babf-1b4c298c890b
sockets: 2
virtio0: ceph_stor:vm-105-disk-1,cache=writeback,size=50G
virtio1: ceph_stor:vm-105-disk-2,cache=writeback,size=200G

--
Best regards,
Gosha

 

[spirit]

Hi,
it could be interrupts related.
This kind of cpu usage is not show inside the vm.

do you have sqlserver installed ? It's known to do lot of interrupts
Also ceph access can use lot of cpu if you do lot of iops

 

[gosha]

Hi,
it could be interrupts related.
This kind of cpu usage is not show inside the vm.

do you have sqlserver installed ? It's known to do lot of interrupts
Also ceph access can use lot of cpu if you do lot of iops

No, sql-server not installed. This Terminal server (rdp) for diskless clients (thinstation via pxe). MS Office and web only.
On Proxmox VE 3.x and 4.x with old GUI - works properly.
This happened since the last update...

 

[spirit]

can you try to install on host :

#apt-get install linux-tools

then use the command

#perf top

to see system calls (send me the result).



Another thing to test:

add to /etc/default/grub

GRUB_CMDLINE_LINUX_DEFAULT="quiet transparent_hugepage=madvise"

#update-grub

then reboot

 

[gosha]

...
#perf top
to see system calls (send me the result).
....
GRUB_CMDLINE_LINUX_DEFAULT="quiet transparent_hugepage=madvise"

Ok.
perf top:



And adding the parameter "transparent_hugepage=madvise" to kernel has no effect...



And it could hardly affected because on Proxmox VE 3.x and 4.x with old GUI - works properly.

--
Best regards,
Gosha

 

[spirit]

Seem to be clock interrupts related.
Can you try to boot on previous 4.2 kernel to compare?

 

[spirit]

Hi,
other users have reported the problem on the mailing list.
Proxmox devs are going to investigated this