Seems you are right. Bu very interesting. I have 2 hosts connected to one UPS. I reviewed situation with power, yes, there was switching to ups for a half a second. the whole machine is working, but only SSDs are felt it...
Now i permanently disabled IOMMU in BIOS. will see more time. for few...
Replace its a bit harder, but i have 3 hosts.
There is different power supply units and CPUs (Ryz 7 3900X,Ryz 9 3950X, Ryz 9 5900) and on all of them the same problem.
Hosts are not loaded high (avg cpu usage 10%, memory not higher than 60%).
Trese is no any usb devices.
That is a host with only power and network cables connected))))
I tried to disable iommu in bios. Still no effect.
The worst thing that i cannot see what causes this.... It works fine, but in some moment it happens. And nothing in logs
Actually, i do not need IOMMU at all. I'm not using devices passthrough to VMs. i tried to passthough GPU but there is nothing good with this motherboard. so i removed "amd_iommu=on" from grub.
also there is nothing connected to usb ports.
I can tell you more))) I'm using Linux not so long (~10years) but i didn't ever seen that some of disks (new hardware) suddenly was lost. I posted in in my separate thread. If interesting https://forum.proxmox.com/threads/storage-lost.87189/
My hardware config is very simple. 1NVMe with proxmox installed and 3 SSD in SATA ports. With replication enabled i didn't try.
But very interesting thing that neither nmap nor in logs i couldn't see what causes this.
Maybe offtopic, but recently i faced the same problem.
But all storages are LVM-Thin. For many tries, reviewing logs (there is nothing interesting) i found that if in VM disk "discard" option is enabled - the problem exists. Simply disabling discard in VM which needs to be migrated - all works...
No. there is very simple host config. NVMe is on M.2, on rest 4 SATA is 1 HDD and 3 SSD.
All SSD is new. (on one host Goodram CL100, Crucial MX500; on other 2xAMD Ryzen5 and Sams QVO 870)
I found interesting post https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1610622 here, but disablin...
Time by time i have a problem that my SSDs have an errors an storage is unaccessible.
A few lines from log:
And then with all othed SSDs.
But NVMe (on which proxmox is installed works fine).
After host restart all works fine.
Same problem i have on few nodes (on other proxmox...
I mean that E5-2630v4 have 25Mb cache but reported only 16Mb.
"args: -cpu "
replaces all cpu flags. So consider to use "host" then view
"qm config VMID --verbose" and use all cpu flags from there and only then add -hypervisor, +ht (or +svm for AMD)
But on AMD processors -smp threads...
thank you a lot.
it works great.
But now i have a different problem. i'm doing it to make users to do "clean" OS install. OS images i don't want to do unattended.
So i need to remove all data in vm image, maybe format it and then start OS install.
Tried to use...