Recent content by DamienDye

"pcie_aspm=off" ??

you can also use the vmware pvscsi option to emulate there scsi controller to so the VM should not know its running on KVM in proxmox rather than VMware

Building on what Francois asked -- that error "unable to get monitor info from DNS SRV with service name: ceph-mon" is a DNS resolution issue. When you don't specify the cluster or monitors explicitly, the Ceph client library tries to auto-discover monitors by looking up DNS SRV records for...

Check your VM is on chipset Q35 if not change it to Q35. Your lspci show your drive MaxPayload is 128 bytes at the moment but the drive supports 512. so lets add "pci=pcie_bus_perf" to the vm kernel command line ASPM policy is default try setting it to performance or adding "pcie_aspm=off"...

Can you post the actual Windows build numbers for the affected VMs? Run winver or check Settings > System > About inside the guest. The specific build matters because Microsoft changed how VBS and Hyper-V enlightenments behave in certain updates, and that lines up with the kind of CPU spike you...

Have you done owt inside the guest to tune Windows for running in a VM? Things like disabling the visual effects, killing hibernation and Fast Boot, setting the power plan to High Performance, deferring TRIM, and stripping out the background I/O that Windows does by default? Without a real GPU...

This issue is not specific to Proxmox. Microsoft have acknowledged the same problem on their own platforms. On Azure, VMs with VBS enabled stopped booting after the July 2025 updates. Microsoft released out-of-band patches KB5061977 and KB5064489 to fix it on Hyper-V and Azure. On VMware ESXi...

Run this on the host to check which NUMA node each NVMe sits on: for d in /sys/bus/pci/devices/*/nvme; do echo "$(dirname $d): NUMA $(cat $(dirname $d)/numa_node)"; done Then check which NUMA node your vCPUs are pinned to -- if they don't match, that's your problem. What CPU type is set on...

you need to either use 2MB pages or 1GB pages not mox like your trying to do. huge pages will only be used on VMs you have set to use huge pages with else they will use standard 4k pages. huge page allocation also needs to take into account numa so you will have a huge page slab per numa node...

hi guys search didn't bring me a result for this but am having issues with ram and CPU hot plug. in version 9.1.4. the proxmox UI doesn't apply them, despite hot plug being enabled, it doesn't appear to work on any VM regardless of guest OS. can somebody tell me why this isn't applying as per...

I would recommend running ceph on it's own NIC's at least 40gb and NVMe if your IO heavy so that you have enough bandwidth for the ceph inter OSD traffic. as for every 1gb there is 2 x 1gb being generated for the additional replicas to be made.

In my current role for production sizing of ceph for customers I recommend 6 nodes with and ec of 4.2 which is still 66% usable you could run 3.2 with 5 servers at 60% usable. If you could add 2 servers you would get the resilience you need. The production cluster I had at 2.1 encoding is...

I started at my last employer with 3 nodes then expanded to 7 once we was passed the 12 month POC stage. it works fine, but you can only ever have 1 server out taking 2 out will cause the ceph to take all the ec pool placement groups offline. I had an employee that did it twice when she was...

the max you can do is 2+1 to give 0.66% efficiency and this is your only EC option since you only have 3 servers.

this is fixable with persistent network naming that uses mac addresses and works with any linux distribution using systemd add "net.ifname-policy=mac" to /etc/default/grub on GRUB_CMDLINE_LINUX line update-grub systemctl enable systemd-network-generator.service if your systemd is new enough...