Recent content by Curt Hall

This started for us this week, all our PVE's are 9.1.4. Our PDM alpha and a newly spun up version 1.0.1 neither one will talk to the 3 PVE's now. certificate error? or kernel update/ firewall? not sure what's happening. EDIT: we (my tech) had wrong password, PDM 1.0.1 does connect now, issue...

Not an idea to help you exactly, but more of a be aware; I have a remote office that has a 4ms ping and I can't maintain a cluster. I had to create a remote cluster and watch all clusters with the new Datacenter manager from Proxmox. So connecting over your connection would be something I...

2 suggestions (not that you will implement either but here it goes): 1. Proxmox host should only ever see drives used as VM storage in a RAID format of some kind (hardware based or RAIDZ or something), the higher the RAID the better (EX: RAID10 better than RAID5) 2. If you turn on a feature and...

We had trouble with PBS until we decided to install it on another host altogether (CPU,RAM,HDD) It was nothing but trouble until it was isolated (hardware wise) from the VM's and drives said VM's are on, its backing up. You can install it in Proxmox, but just let that host be for PBS purposes...

Well, first off, you didn't say which NVidia card you had, and I do mean "card" and not GPU and here's why. There are cards that have multiple GPU's on them. Those can be split up to multiple VM's. Let's say its a 4 GPU card ($5k-$10k), then 2 VM's can each get 2 GPU's, or 1 VM gets 3 GPU's the...

I'm not saying that, my SSD's are not consumer. But if you had enough of them and ran RAID10 like mine then the wearout per drive is better the more you add. That's the takeaway here.

Also, I have 16 of them in ZRAID10. The host is running 60 Win11 VM's on those drives for thin client kiosks in our public area.

Samsung MZ1LB960HAJQ-0007 here's smart data on one of them: SMART/Health Information (NVMe Log 0x02) Critical Warning: 0x00 Temperature: 39 Celsius Available Spare: 100% Available Spare Threshold: 10% Percentage Used...

Interesting perspective on SSD wearout (assuming that's what you are referring to). I have a 5 year old server running Proxmox and ZFS on M.2 drives, they are just now at 29% wearout. How long do you expect them run for? I have 10 more years on mine, just curious?

We have AMD Epyc Dell server with A16 card, seems ok for us also. However, the original post was for issues with the consumer grade 5090, which does not play well in these type environments.

I am also in the Proxmox Facebook group (well one of them at least, as I'm sure there's many), and in there people reporting no issues with their old 4090, but when they switch to a new 5090, its nothing but issues. They also don't have solutions either. Seems like a Nvidia issue to me at this...

@frankmanzhu not wrong per se, so don't confuse my point of view with his, I maintain multiple VM's per host (60-100) and cannot afford to have vGPU tripping my host out. but if @simon11118 has only the one or very few other VM's on that host it could be worth a shot, my environment just can't...

Sorry @simon11118 if not helpful, by steering you away from passing through non-vGPU's as if they were vGPU's, but @alexskysilk is right, it would be better on physical box for sure.

Its generally considered not advisable to use a "gamer" GPU card for a passthrough-to-vm AI card. Even it it connects to the VM, its still just a 5090, your AI will be the slowest thing you ever saw. We have done this trying to use various cards, some Nvidia (gamer) style and some Tesla cards...

might be better off with one of these since you need the other peripherals to work also, but you would need type C port on the host