Hi
We are trying to get our H100 GPU passthrough to work again after upgrading our Proxmox host from PVE v9.1.1 to v9.2.3. It worked ok on v9.1.1.
We have set the config for the VM (running AlmaLinux 9.8) on the host to:
hostpci0...
Hello,
I’m trying to set up a 2-node Windows HA / Failover Cluster on Proxmox.
Has anyone here successfully achieved this setup on Proxmox? If yes, I’d appreciate any guidance or best practices.
If only one core is submitting and waiting for IO wouldn't that result in a maximum average iowait of 1/core_count? i.e. at most 3% if 31 cores are idle.
One core can still easily saturate the entire disk bandwidth, so it would be better to...
your pg number seem quite low, it could impact write performance with pg_lock.
try to set pg target ratio to 1.0 in your ceph pool
Also, verify in your bios that your server is set to max performance profile. (you want cpu always at their max...
Odd
Odd, the article you referenced states:
'Many RAID HBAs can be configured with an IT-mode “personality” or “JBOD mode” for streamlined operation'
I've used "JBOD mode" without issues on the past, both for Ceph and ZFS. It disables block...
Have you considered using a gold or silver level Proxmox partner for support? Many will allow you to ad-hoc reach out to them when support is needed for an hourly fee (~$200-$300/hr) without committing to an annual support minimum. This allows...
Unter welchem Windows läuft das denn?
Windows 11 24/25H2 - dann wäre das nur logisch, da die HyperV für ihre eigenen Sandboxsystem nutzen. Da ist nix mehr mit nested.
Zieh das PVE auf einen eigene dedizierte Server.
A dual socket system with 2x Intel Xeon E5-2690v4 has the potential for 72800 "cpu units" of performance (more with turbo+hyperthreading but lets leave that for now)
a dual socket system with 2x Epyc 9575F has the potential for 422400 "cpu units"...
Wie in aller Welt kommt man darauf, PVE nested zu betreiben?
in solchen Fällen, die ich meide wie der Teufel das Weihwasser, suche ich zuerst im System, welches die "Hardware" zur Verfügung stellt.
Sehe ich auch so. Fehler steht ja schon in dmesg drin:
Jun 30 13:39:33 pve QEMU[64897]: KVM: entry failed, hardware error 0xffffffffffffffff
Jun 30 13:41:15 pve kernel: kvm: kvm [68075]: vcpu0, guest rIP: 0xffffffff8388840f Unhandled...
in regards to subscriptions beeing too expensive for the homelab...
i run 3 pve-nodes and one pbs-node in my homelab.
i have a subscription on one of my pve-nodes because i apreciate what the proxmox-team is doing and want to support them.
the...
RAIDz1 is not at all like RAID5 in performance and usable space. And it's indeed not good for VMs. This had been mentioned often, see here for additional information:
https://forum.proxmox.com/threads/fabu-can-i-use-zfs-raidz-for-my-vms.159923/
OK, so I came up with this brief patch against PVE/QemuServer/Cloudinit.pm:
diff --git a/src/PVE/QemuServer/Cloudinit.pm b/src/PVE/QemuServer/Cloudinit.pm
index c1311da8..b3d2e9e2 100644
--- a/src/PVE/QemuServer/Cloudinit.pm
+++...