P
prf
Guest
I am unsure whether this is the correct forum to post this question on but I thought I would start here. Please let me know if you can suggest other more appropriate forums.
I am successfully running a 2 node Proxmox v1.7 cluster for a production environment of Linux and Windows 2008 R2 servers.
pveversion -v
pve-manager: 1.7-10 (pve-manager/1.7/5323)
running kernel: 2.6.18-4-pve
proxmox-ve-2.6.18: 1.7-10
pve-kernel-2.6.18-2-pve: 2.6.18-5
pve-kernel-2.6.18-4-pve: 2.6.18-10
qemu-server: 1.1-25
pve-firmware: 1.0-9
libpve-storage-perl: 1.0-16
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-9
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1
pve-qemu-kvm-2.6.18: 0.9.1-10
I am using OMD v0.44 (nagios + check_mk + mk-livestatus + pnp4nagios and nagvis) to monitor my infrastructure including the virtual servers. check_mk uses an agent on each server to obtain performance data which is then queried by a service on the nagios server.
This is working well on both Linux and Windows guests except I have discovered the Windows network counters for bytes sent and received are zero. For physical servers, these counters are non-zero and increment as expected.
I am using the e1000 network device in the Windows 2008 R2 guests and the default Microsoft driver (Intel PRO/1000 MT v8.4.1.0) for this device within the guest.
The check_mk agent determines Network Performance Interface Performance Object counters (amongst others).
At this stage, I am guessing there is a difference between how performance counters behave in physical versus Linux KVM Windows 2008 R2 environments or the check_mk agent is behaving differently in these two environments?
Could it be related to the KVM, the e1000 virtual driver in KVM or even the physical NICs (Intel 82576 Gigabit) of the underlying virtual server hardware?
Comments and suggestions welcome?
Thanks
Paul
I am successfully running a 2 node Proxmox v1.7 cluster for a production environment of Linux and Windows 2008 R2 servers.
pveversion -v
pve-manager: 1.7-10 (pve-manager/1.7/5323)
running kernel: 2.6.18-4-pve
proxmox-ve-2.6.18: 1.7-10
pve-kernel-2.6.18-2-pve: 2.6.18-5
pve-kernel-2.6.18-4-pve: 2.6.18-10
qemu-server: 1.1-25
pve-firmware: 1.0-9
libpve-storage-perl: 1.0-16
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-9
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1
pve-qemu-kvm-2.6.18: 0.9.1-10
I am using OMD v0.44 (nagios + check_mk + mk-livestatus + pnp4nagios and nagvis) to monitor my infrastructure including the virtual servers. check_mk uses an agent on each server to obtain performance data which is then queried by a service on the nagios server.
This is working well on both Linux and Windows guests except I have discovered the Windows network counters for bytes sent and received are zero. For physical servers, these counters are non-zero and increment as expected.
I am using the e1000 network device in the Windows 2008 R2 guests and the default Microsoft driver (Intel PRO/1000 MT v8.4.1.0) for this device within the guest.
The check_mk agent determines Network Performance Interface Performance Object counters (amongst others).
At this stage, I am guessing there is a difference between how performance counters behave in physical versus Linux KVM Windows 2008 R2 environments or the check_mk agent is behaving differently in these two environments?
Could it be related to the KVM, the e1000 virtual driver in KVM or even the physical NICs (Intel 82576 Gigabit) of the underlying virtual server hardware?
Comments and suggestions welcome?
Thanks
Paul
Last edited by a moderator: