Thanks! Just needed to be 200% sure!

I'm not entirely sure it could work without having serious issues. Maybe nested will be in better shape in 8.1, but it's a known 'thing' that nested virt code is a bit fragile at the moment.

@MajorTom thanks that could be a solution. In the end I moved my single-machine Homelab to Proxmox, because I have way better control in this specific use-case - I can see IOMMU groupings and figure out which USB controller I can also passthrough. And this works fine, I'm posting this from this VM. XCPng isn't bad, it´s just not suited to my very specific use case. Cheers, Clément

Right, if your used is 123.4/128, there seems RAM crunch to migrate anything around 4GB. Can you try migration after freeing up some RAM? Also what version of XCP-NG is being used? There is a known issue related to migration - you can find it on forum.

AMD keeps things simpler for security and performance impact of spectre patches. AMD offers just as much performance in a simpler GPU package. AMD for CPU and AMD for GPU. Thank You for all your work @olivierlambert

Thanks! Enjoy XO and XCP-ng

Can you be more specific? I think you are mixing number of CPU inside the machine and visible CPU inside the dom0. Remember, Xen is type 1, the "host" you are connecting into is not the physical host, but a privileged VM.

@rjt The November 2019 security, bugfix, and enhancement updates address this issue in the section "Enable increasing max_grant_frames for guests." Indicates that with not enough grant frames, the guest VM will not be able to use the all the virtual NICs that it expects. Still wonder if increasing max_grant_frames would speed up existing network interfaces even when they have enough working vNICS.

@andersonalipio Since something is up with the API, restarting the xapi toolstack on the host might help. Do not try migrating the running VM at first, but move to the shared NFS StorageRepository. Then after the VDI is hopefully successfully moved, then try moving the running VM. Does it move to the other Citrix XS 7 host?

During VM export, the ssh connection is gone and the result is frustrating. the disk of the virtual computer does not appear at all. Xcp-ng 8.0 use. I think he sent the disk to eternity.

The message was relatively clear (for a XAPI error This operation cannot be performed because the specified VDI could not be found on the storage substrate sr: 28a8b360-7b1b-58ca-81b6-f907c164ec38 (XCP-ng Tools) vdi: b1a7a49a-f17c-42ac-af73-4429eb57b07a (Old version of guest-tools.iso) This means the VM can't boot with this current CD because it's not there anymore. I suppose you updated your host without ejecting all CDs before. So the old tools CD was removed for the new one, and your VM kept it. So when you tried to boot, it couldn't because it couldn't find the old ISO.

Thanks Olivier. Yes I am able to set the parameter now using xe commands. Thanks for your help!

One socket right?

Just courios - what servcies do You still have on Server 2003?

@KFC-Netearth Welcome

Then I guess it's empty. I guess I'll just redo the server, and chalk this up as a one in a million file corruption, for which we didn't have the proper OS backup to counteract. We do have backups of the VMs anyway. Thanks for the help! We didn't get this fixed, but I did learn a lot! This is our first Xenserver/XCP host.

@olivierlambert That's true...

If you know exactly what's checked in the driver (CPU vendor, ACPI tables, BIOS info, SMI?), we might be able to dig into the right direction. It's just that I won't spend a lot of resources without knowing if our modifications are useful or not.