Recent content by pcravero

Having contributed in the past to that specific wiki section, I can say that all those steps can be applied to VMware Linux guests. Just leave out Micros.ft specific tasks (as a Linux sysadmin you should easily recognize them ;)). As an outcome of this thread I have added a short note to the...

As Udo suggested, your VMs on the 10.x LAN will have their own network configuration, including a correct gateway for their subnet. You don't need an IP on eth1/vmbr1, unless you want to be able to access PVE Web console on the private IP instead of the public one. Paolo

Hi. That looks like a standard configuration for a host. All traffic goes to the default gateway on eth0 except to 10.7.6.201/255.255.255.224. So if you are pinging to/from 10.7.6.1 you're most probably not getting a response. Can you please specify: - the virtual machine IP address - the...

Hi. I assume your multihomed machines have 2 vNICs and all VMs are on the same physical host. Also that VM1 has 10.0.0.2 not .1. How about DNS name resolution? A slow to start SSH connection (on the 10.x net) reminds me of a timeout within hostname resolution. Paolo

Your Proxmox server sees two NICs: eth0 and eth1. eth0 should already be configured. Proxmox has probably associated a bridge to it: vmbr0. Log into the web console and create a new bridge device (Configuration > System > Network), giving it an IP address on the FTTB network. Call it "vmbr1"...

VMware to PVE/KVM We had to migrate a VMware ESXi 3.5 to PVE 1.8/KVM. We first tried to re-use directly the .vmdk (originated from a clone on VMware) but it didn't work. We then tried to convert to qcow2 with the supplied command qemu-img convert -f vmdk myImage.vmdk -O qcow2...

Quick self-reply. When upgrading to 1.8 I moved the running VM to another cluster host, and there it showed the correct CPU load. Moved it back to the original host still reports correct CPU load. What went wrong is still TBD.

Hi. I have been running for 26 days a KVM CentOS 5.5 with 1 CPU/core. I made the VM run the CPU at 100% 24/7 (john the ripper with an impossible task). Today I noticed that the WebUI reports 0% CPU utilization for that VM, while until 5 days ago it was correctly shown at 100%. The process is...

Maybe it is in the docs and you've already discarded this procedure. This is a workaround to the missing kvm template feature, not an answer to your question. Prepare a working VM and take a backup of it with proxmox. Then restore it using a different VMID, remove and re-add the NIC(s) (to get...

Re: SCSI or Virtio Driver Issue: Windows Server 2008 R2 64 Bit [SOLVED] Sorry to bring up a somewhat old thread, but my question is related: is Windows 2008r2 able to use PVE/KVM SCSI disks? I can mount IDE disks. I know the trick for virtio, but SCSI? Strangely enough, Windows 2008 NOT r2 can...

Bingo! For the records (and the less experienced in proxmox): modify /etc/qemu-server/<VMID>.conf . Make the bootdisk: value match the actual disk label (ide0, scsi0, virtio0, ...). Refresh the vmlist.htm page and the correct size is shown. Nothing has to be rebooted. Cool. Paolo

@dswartz, not really! I set virtio on all Linux VMs and it always seems to work, except if I am re-using a VMID (but not 100% sure about this, might be a coincidence) See:

Hi. I'm running: ~# pveversion -v pve-manager: 1.7-11 (pve-manager/1.7/5470) running kernel: 2.6.32-4-pve proxmox-ve-2.6.32: 1.7-30 pve-kernel-2.6.32-4-pve: 2.6.32-30 qemu-server: 1.1-28 pve-firmware: 1.0-10 libpve-storage-perl: 1.0-16 vncterm: 0.9-2 vzctl: 3.0.24-1pve4 vzdump: 1.2-10 vzprocps...