![[FS][EU] NVIDIA P2000 5GB vRam 80 EUR + Shipping](https://external-preview.redd.it/Kbp5DYxNB-QpTkfOjhVWSNxLGy1YfaROn4DzEPw21hk.jpeg?width=1080&crop=smart&auto=webp&s=bd9233c90cf9bc752cfb96a3ebea39f599fb1bb1)
[FS][EU] NVIDIA P2000 5GB vRam 80 EUR + Shipping
https://imgur.com/gallery/p2000-nvidia-BhNFzq5
NVIDIA P2000, I ship same day from Italy if you order before noon.
Pulled from Dell Workstation.
![[FS][EU] NVIDIA P2000 5GB vRam 80 EUR + Shipping](https://external-preview.redd.it/Kbp5DYxNB-QpTkfOjhVWSNxLGy1YfaROn4DzEPw21hk.jpeg?width=1080&crop=smart&auto=webp&s=bd9233c90cf9bc752cfb96a3ebea39f599fb1bb1)
https://imgur.com/gallery/p2000-nvidia-BhNFzq5
NVIDIA P2000, I ship same day from Italy if you order before noon.
Pulled from Dell Workstation.
Looking for an Xenon E-2378, preferably HPE branded or unbranded.
See title
<Steve> Anybody knows what the red cable is for?
<Dan> Connection terminated (timeout)
<Amy> Connection terminated (timeout)
<Robert> Connection terminated (timeout)
<James> Connection terminated (timeout)
<Brian> Connection terminated (timeout)
<Jason> Connection terminated (timeout)
<Mike> Connection terminated (timeout)
<Henry> Connection terminated (timeout)
Propably it's me, but by Now I ran out of ideas.
When trying to update cold spare Microservers from Intel SPS 06.00.03.200(B) (link to HPE Download Page) the update fails with Post Code: 0003B. This occures despite having updated the system rom to >2.60 as required. Also intermediate steps failed with identical symptoms (e.g. from 06.00.03.200 to 06.00.03.204). Longest wait I tried was overnight.
Further analysis could not be made as no more logs were created during the process. Attempts to proceed via SPP ISO yielded the same result. Complete removal of all PCIe Devices did not improve the situation.
06.00.03.200
Current component firmware version(s):
Screenshot: https://i.ibb.co/Psg5RHpH/Screenshot-2026-05-10-at-21-16-17.png
UPDATE:
HPE does not support this feature on all devices and, of course, doesn't advertise on which servers online firmware upgrade should work or not work. IP Has silently been deprecated for some Gen 10 plus models (without any list) IP online firmware upgrade support.
F*** you, HPE!
In general (and this goes also for you, Dell) I keep struggling gasping the difficulty of configuring
What I see is a tendency, especially on HPE, to change the logo in every second firmware upgrade, but not being able to connect my provisioning tools to a vendor owned server. What I found though were literally 20 different support responses on forums treating exactly the above described topic with replies that didn’t work. My first contact with HPE reminds me while in the company I switched to Dell….
I would like to offload every and all logs to external journals. Is that even possible? I have some legacy systems running on esxi that I would like to migrate to proxmox, the physical infrastructure requires a lot disk space and replacing the current boot media would require a lot of hardware reconfiguration on services which are redundant (hence we can accept a higher failure rate on hypervisor boot media), we have discussed this internally and decided we want to give proxmox on industrial USB sticks a try for science given that the systems are not mission critical.
So here are my questions:
- which logs need to be redirected?
- how persistent is this change regarding to proxmox major updates?
- has anybody done this?
Proxmox 9 (Debian Trixie) is not officially supported by the HPE Ilo management agent. You can make it work though by downloading and installing the Package directly and installing it manually. Keep in mind that the agent for ILO4 (AMS) and ILO5 (AMSD) are not the same. The ILO4 agent will refuse to launch on ILO5 systems (= Gen 10 including plus and v2).
root@pve:\~# wget https://downloads.linux.hpe.com/SDR/repo/spp-gen11/2026.03.00.00/packages/amsd\_4.6.0-2131.3-ubuntu22\_amd64.deb
root@pve:\~# apt install ./amsd\_4.6.0-2131.3-ubuntu22\_amd64.deb
root@pve:\~# systemctl status amsd
● amsd.service - Agentless Management Service daemon
Loaded: loaded (/usr/lib/systemd/system/amsd.service; enabled; preset: enabled)
Active: active (running) since Mon 2026-05-04 12:24:23 CEST; 43s ago
Invocation: 4ca3607a2c644334a5d721c4f37ac2e2
Main PID: 536684 (amsd)
Tasks: 1 (limit: 34876)
Memory: 2M (peak: 6.3M)
CPU: 54ms
CGroup: /system.slice/amsd.service
└─536684 /sbin/amsd -f
May 04 12:24:22 pve systemd\[1\]: Starting amsd.service - Agentless Management Service daemon...
May 04 12:24:23 pve amsd\[536684\]: amsd Started . .
May 04 12:24:23 pve systemd\[1\]: Started amsd.service - Agentless Management Service daemon.
Some of you might know from writing on of the most read posts regarding the migration from vmWare to Proxmox on Reddit with almost half a million views, my scripts to deploy SSL with let's encrypt on Windows Servers or my disk formatting guide and scripts with thousands of readers and users every year.
As many of you I grew up with the 8th Gen of the HP Microserver and I always saw it as an average server with a good IPMI solution and some painful limitations (16 GB Ram) in a well engineered form factor. HP's compromises on this platform have always been painful, but the physical usability of the case made up for it in many aspects. It's like the Volkswagen of Homeservers, a little luxury, good build quality and some compromises for a sticker price that was only permissible because you hoped it would run for ages. As many of you I use decommissioned hardware at home, which means for many of us G8-10 HP or Gen 12 to 14 Dell Systems, with all the drawbacks and benefits.
So in recent years two major things happened in IT:
, which can be downsized to one slot.
As some of you might know, vGPU is far from an easy implementation on system level requiring a lot of tinkering and funny enough, requires some features that seem hard to come by on HP and Dell servers:
Both features are crucial if you want to do anything but pass a GPU directly to a VM hence fractioning a GPU into two or more virtual GPUs (or "vGPU"). Both are, extensively part of the UEFI standard, even if you might not see the features, because the manufacturer hides them, odds are, they are there. Dell and HP servers are special beasts though, the UEFI is not made by a 3rd party, but by Dell and HP themselves giving them larger leeway to suppress or enable features, that other manufacturers just hide.
So here is my plan to get vGPU on Microservers running:
But I need your help first:
To safe me from the pain of wasting 1200 EUR on hardware, I need to know one thing:
does the UEFI of Proliant 10th gen servers support 4G Decoding. Unfortunately HP encrypts the update BINs so it's not possible to figure this out without having a physical machine.
What will you need to help me out on this:
What will the comunity get back?
vGPU using Intel Arc and Proliant servers on Proxmox.
What needs to be done?
In Short: read out the BIOS and tell me if the configuration options to make vGPU on this system possible are only hidden or non existing.
It's fully documented here: Enabling hidden 4G decoding
So who is nice enough to give this a try?