The original ideas are from:
Hi Dominik. Thank you for the reply.
1. I meant the blacklist.conf and vfio.conf were copied to both kernel.
My goal is to only apply those two to one of the kernel.
In an event that I bricked the latest kernel, I can still boot to the other kernel to make changes (like ip address).
1. Because I passthrough the only graphics card, once the PVE gets to "Reading all physical volumes...", I no longer can use the console.
If I make some change that lock myself out of web interface (for example, change network config), I guess that I have no way to get into the system...
Due to some restrictions, I have to use two NICs on my Proxmox node in the following way:
NIC1: 10.10.10.xxx, just one IP assigned by the facility to the Proxmox Node, able to access internet via their firewall
NIC2: 192.168.150.xxx, connected to my own router (192.168.150.1)
Like stripe Disk 1 2 3 <-mirror-> stripe Disk 4 5 6?
I probably didn't (I used all defaults), is there a way to check it now?
Yes. These I know. my total capacity utilization is 11%.
Four SSDs are in Asus Quad Card with fan, at 35C. The other two are under motherboard heatsink, which is a...
PVE: Latest 6.4-5 (PVE 5.4.106-1)
Total system RAM is 256GB
SSD is WD SN750 2T x 6 in RAID-Z2, compression=lz4
VM : cache=writeback discard=on, size=501G, ssd=1 (see attached code)
VM Crystal Disk's RND 4K and IOPs are much slower than individual drive in bare-metal system.
Read a lot and final find a working solution.
When it gets to start menu GRUB.
1. Edit grub command line (press 'e' key )
initrd=\EFI\proxmox\5.X.X-X-pve\initrd.img-5.X.X-pve BOOT_IMAGE=/boot/linux26 ro ramdisk_size=16777216 rw quiet splash=silent (Or whateever is there)
Urgently need help. I have research work data on it:oops:
My GPU passthrough has been working well on both 6.2.55 and 65. Because I used update-initramfs -u -k all, both .55 and .65 has GPU frame buffer disable thus no host display out. I was trying to reverse GPU passthrough...
I have an old one-slot Quadro card that I can use.
Or I can buy a "ZOTAC GeForce GT 710 1GB DDR3 PCIE x 1" in slot 4.
(that way I can still use slot 2 for 10G NIC and slot 8 for 2nd GPU for deep learning on slot 5 with PCIE riser)
Currently, I have a 2080Ti on PCIE slot1 (x16) and ASUS Quad...
My motherboard does not have serial COM port on the back panel, nor does serial COM header on the board.
Is that something that can be added using USB-serial COM or PCIE x 1 serial COM card.
I have other computer to use. Even a very old one that has serial COM on motherboard back...
I had some issue to connect to my proxmox sever which does not have IPMI (or iKVM).
It is build on a Threadripper workstation with MSI TRX40 motherboard.
I fixed the issue, which is not related to the proxmox.
That lead me thinking:
Since my only GPU has been passthrough, the frame...
Thanks for the very detailed reply.
This server is a somewhat proof of concept and a relatively affordable one to show my boss the capability of using Virtualization to utilize the high core count threadripper's performance. With lesson learned on this one, we might consider Lenovo P620 with...
I have created two zfs pools with the simplest command: zpool create <pool> <device> at the begining.
It seems have no log (aka zil or slog) and cache. I believe I did turn on lz4 compression for all three.
My question is: in my case, do I need log or cache for each pool?
using GUI, at LXC700 page, restore function
I attempted to do a restore of LXC700 from a vzdump, which I know was working.
It end up with an error and my LXC700 completely disappeared (nvmepool/subvol-700-disk-0 disappeared).
The mount point mp0 (my SMB data volume) at...
The turnkey LXC is unprivileged.
Downloaded from Proxmox template search: debian-10-turnkey-fileserver_16.0-1_amd64.tar.gz
Please help, I have important data on the LXC.
Two days ago, I rebooted host, and Turnkey SMB LXC doesn't start due to this error.
I searched around and...
But the baremetal Win10 benchmark that matlab is giving as an Threadripper 3970x example doesn't seems to suffer from this performance issue.
Also the VM500 (the one with GPU passthrough is also fine).
So what could be different from VM500? From VM config, I have set the CPU argument the same...
Thanks for the reply.
No they are not running at the same time.
The only other thing running is a Turnkey SAMBA server in LXC container using 1 VCPU and 512MB RAM
CPU: AMD Threadripper 3970x (32C/64T)
GPU: NVIDIA 2080 Ti Founders Edition Blower Style
MB: MSI TRX40 Pro Wifi