Search results

Seems nested virt. active by default since Win11. Try 'max' vcpu type : cpu: max args: -cpu max,-hypervisor

"host" vcpu type for Win11/Win2025 guests is slower than x86-64-v2-aes/v3 ( best is "max" ) because nested virtualization is often active because of VBS, where other vcpu type prevent nested virt.

No problems and reliable here with two small cases for Sync Job over SMB over WAN , a not optimal path and not recommended but works fine and speed is honorable : 1/ since 3y : Datastore over SMB over WAN, Sync from local Datastore USB HDD ( ~2H daily for 15~20GB pulled said in summary sync job...

EDIT: Dit you add back the Datastore in the PBS side ? with "Avdanced" tick to reveal "Reusing existing datastore"

Fleecing mitigates, but the recommended and safest is local BPS backup then Sync over WAN/VPN. Fleecing from the docs : When a backup for a VM is started, QEMU will install a "copy-before-write" filter in its block layer. This filter ensures that upon new guest writes, old data still needed for...

Can be your http application only accepts LAN IPs

https://www.proxmox.com/en/services/training-courses/videos/proxmox-virtual-environment/whats-new-in-proxmox-ve-9-1 typo "for Kernel 16.7"

yes

"max" should be the best performance. "x86-64-v2" without aes is for ancient cpu, what is your real cpu ?

"host" is not more the fastest since Windows latest version often enables virtualization which triggers the unwanted nested virtualization.

post screenshot of the Controller tab. No RAID does not meant no RAID controller.

Don't forget to enable "Max Performance profile" in BIOS and double check if Disk Writecache is enabled, often it's disabled because designed for RAID embedded cache usage. BTW, I repeat, try with Lvmthin before all the tuning about ZFS. It could be that your problem is elsewhere.

You can try "max" CPU type

Try disabling Windows Firewall "Remote Desktop" UDP-In rule to dig out.

I don't see the relationship with the Agent. is PVE firewall enabled ? is RDP over UDP used ?

Are you sure isn't the clouflare tunnel worse performance and more latency ?

Do not use "host" CPU type in your VM. You can try "max" if no Live migration used. Or model close your physical host CPU.

Doesn't boot either with EPYC-Rome-v4,-hypervisor where EPYC-Rome-v4 alone works. Works with args: -cpu EPYC,-hypervisor Waiting the next month if Microsoft fixes the fix, like the Stop Code UNSUPPORTED_PROCESSOR BSOD.

Try your SQL under Lvmthin during a week, to exclude ZFS or not. CDM benchmark show Fujitsu on ESXi is slower than Dell on PVE.

Seems I'm encountering a bug. Host CPU is AMD 7302P ( EPYC Zen2 "Rome" ) on HPE DL325 G10. PVE 8.4 ( Enterprise version 8.4.16 ) proxmox-ve: 8.4.0 running kernel: 6.8.12-18-pve pve-manager: 8.4.16 pve-qemu-kvm: 9.2.0-7 qemu-server: 8.4.5 VM guest : Windows 10 ESU, 2019 , 2022 , 2025 , UEFI (No...