Latest activity
-
JHi Daniel, thanks for the reply! not right now, no. I was just thinking ahead to the future :D Right now it is only active on two small 3-node test clusters, one of which has some Nvidia GPUs with NVAIE (yes, works the same as "normal" vGPUs...
-
JThanks! Done: https://bugzilla.proxmox.com/show_bug.cgi?id=7557
-
JIt depends, and it depends what you look at in the Proxmox web GUI. Total and Host Memory Usage are now at least 100% in the graphs for VMs with passthrough. The Used (and the Memory usage bar at the top) depends on what the ballooning driver...
-
JPCI(e) passthrough requires all VM memory (which can be a little over 100%) to be pinned into actual host RAM (due to device DMA). Ballooning cannot give memory back to the Proxmox host, but the ballooning device can (for some operating systems)...
-
CCookiefamily replied to the thread Questions about the dynamic CRS.Hi Daniel, thanks for the reply! not right now, no. I was just thinking ahead to the future :D Right now it is only active on two small 3-node test clusters, one of which has some Nvidia GPUs with NVAIE (yes, works the same as "normal" vGPUs...
-
fba replied to the thread Restoring node, pinned nics.If it is a new node it will have different mac addresses for the nics and these files won't help without updating the mac addresses they refer to. Might be easier to run the pve-network-interface-pinning generate command instead. -
NI need to restore a node; copying what I need from rescue mode. I have all my nics pinned, do I need to backup/restore the .link files under /usr/local/lib/systemd/network? The interface files lists the pinned names so I'm guessing I need to...
-
It depends, and it depends what you look at in the Proxmox web GUI. Total and Host Memory Usage are now at least 100% in the graphs for VMs with passthrough. The Used (and the Memory usage bar at the top) depends on what the ballooning driver... -
LLeFred replied to the thread what the best way to use specific transport on specific header detection ?.An old post of Stoïko said to pass by smtpd:10025 ; trying...
-
jsterr replied to the thread MEMORY PROXMOX 9.0.Thanks! Pinned to actual host-ram means the vm also uses 100% on vm-summary? Or is it "only" reserved on host summary, so vm could theoretically only show 50% memory usage, although it reserves 100% on hypervisor-level?
-
PCI(e) passthrough requires all VM memory (which can be a little over 100%) to be pinned into actual host RAM (due to device DMA). Ballooning cannot give memory back to the Proxmox host, but the ballooning device can (for some operating systems)...
-
TTheGoat reacted to fiona's post in the thread [SOLVED] Applying pve-qemu-kvm 10.2.1-1 may cause extremely high “I/O Delay” and extremely high “I/O pressure stalls”. (Patches in the test repository with
Like.
@martin_comp Yes, the issue affects IO delay (the name for iowait in the UI) too. A fix was applied in git master and the new pve-qemu-kvm=10.2.1-2 package is currently going through internal testing. -
jsterr replied to the thread MEMORY PROXMOX 9.0.Hello Impact. Do you know if Balloon = 1 is ok, when memory max and min memory have the same value (so no actual "ballooning" is done) regarding PCI-PT? https://pve.proxmox.com/wiki/Dynamic_Memory_Management#Ballooning says Ballooning should...
-
TToeiRei replied to the thread Cloud-Init.Ich hab' mal bisschen debugged: Debian verwendet keinen Netplan sondern einfach das Traditionelle ifupdown system (/etc/network/interfaces) oder via NetworkManager (nmcli oder Profile). Doof dabei ist nur, dass cloud-init halt selbst bei Debian...
-
fireon replied to the thread Cloud-Init.Konnte ich so noch nicht beobachten, aber ich habe auch noch kein Debian damit getestet. Vielleicht hilft dir da was weiter: https://forum.proxmox.com/threads/cloud-init-f%C3%BCr-linuxvms-ganze-einfach-unter-proxmox-ve.181926/ Was verwendest du... -
DDenary replied to the thread ha-manager and VM's with host passthrough.Hi dakralex Thanks for the response. I'll add myself to the mailing list for the bug report. If I give the VM a host preference, does that make it the last VM to migrate from a node for the new dynamic scheduling feature? Thanks
-
Siehe https://forum.proxmox.com/threads/proxmox-backup-server-4-2-released.183130/
-
HHydrant1512 replied to the thread Proxmox VE 9.1 / ZFS 2.4.1: orphaned zvol destroy clears error temporarily, but txg_sync hangs in D state and reboot resurrected the orphan.That is consistent with what I am seeing.So far this looks isolated to a single orphaned corrupted zvol (servertank/vm-200-disk-1), not a general read failure across the pool. The mirror members are still ONLINE with 0/0/0 ZFS counters, and both...
-
jsterr replied to the thread Questions about the dynamic CRS.Thanks! Done: https://bugzilla.proxmox.com/show_bug.cgi?id=7557
-
Yes, good idea! We already thought about this during development, though we were focusing on the core feature first. Feel free to create a Bugzilla entry [0] for this in the mean time, though as this is relatively trivial to implement and as you...