@stormi I’ll be back on Wednesday (just short holiday now), I’ll try your advice and see how it works

@TheRedHerring I have an HM90 that I use as a host and it worked with 8.2 out of the box except for the 2.5G NIC. I switched to 8.3 and everything worked out of the box.

As a server virtualization platform, ECC is the most common memory used by our XCP-ng deployments all around the world It's up to the system (CPU/memory) to deal with it and correct errors on the fly, it's transparent for the hypervisor

That's weird, never heard about this option to be needed to make it work, does it ring a bell @fohdeesha ?

No worries. There's some politics involved also to be sure it will be accepted upstream, but we expect to be a matter of months (at worst) to be upstream, and then it should be straightforward to get it into 8.3 Because on XO side: it's already working! (we are using our own patched XAPI to make the tests)

Thanks everyone! I’ll let you know if we have any issues.

I would run it in Scale if i were you. Ive read about people's set ups like this and there hasn't been any issues with it but It has been argued that virtualization of your storage underneath a hypervisor is dangerous. I don't know what type of pitfalls you might encounter with this but it will surely make recovery from a disaster a little more complicated. And with Scale being able to run VM's you can have the GPU passed through to your Win11 machine. I presume you are transcoding in your Jellyfin? I use Plex and I pass through a GPU to it (but thats in esxi). Can you pass through your GPU to multiple devices to use for that? I know there is vGPU but I don't know if it can be used for transcoding. It would be interesting to try though

@gr85z said in New server /storage setup to replace current systems: A couple questions on setup. Does anyone know if there is a guide on setting up something like this? Also with XO can you run 2 XO instances and have them synced to each other in case one goes down you have an alternative one to use? You'd want to setup XO Proxy to manage separate hosts in disperse locations.

Just wanted to say thank you for this thread. I am in the midst of evaluating ESXi alternatives and trying ProxMox as well as XCP-NG. I just installed XCP-NG this morning and have been trying to figure out if I should be doing hardware RAID vs ZFS raid. This thread provided the answer I needed. Hardware Raid on my Dell Power Edge Servers is easy and reliable. Thank you again!

@ajzride Gotcha, at least restoring is pretty easy. Passthrough is pretty straightforward, just be sure to passthrough the entire device including audio sub devices and it should "just work", been very reliable for me.

@olivierlambert Was never able to get any information out of AMD, so I bought a used FirePro S7150x2 and it works as anticipated, so the issue does appear to be the RX550 does not support mxgpu.

@ajpri1998 Having centralized storage sounds dangerous from the perspective of failure. (We already have a 2PB all flash Vast storage.)

Might be worth installing netdata plugins into your hosts. https://xcp-ng.org/forum/topic/2288/netdata-package-is-now-available-in-xcp-ng Installing Netdata manually ...Just run yum install netdata-ui on each host you want to monitor and that's it. You can just go on http://host_address:19999 and view the different metrics. If you want a centralized netdata and stream everything to a "central" host, feel free to read Netdata documentation on how to do so!

@drewq said in Used server suggestions: @ajpri1998 Thank you for the input. I am kind of partial to the dell power edge series myself. Sounds like that would be the best option going forward. Also aim for Gold or Platinum (preferably Platinum) efficiency PSUs where possible. Taking into account the wattage of the PSU and the hardware's requirements and also considering whether the PSU will be upgraded for future hardware growth. The higher efficiency hardware will mean that it will use less power when in operation. The efficiency ratings range from No Rating (or a minimum of Bronze) up to a rating of Platinum (the highest). Checkout this https://breadnet.co.uk/ as its a fellow home lab user.

@Andrew Thank you for your valuable experience and suggestions. I will personally test it and provide feedback on the results after testing.

Yes, because the CPU features are assigned on VM boot (you can't remove a CPU instruction while the VM is running, as you can imagine, if it's used anywhere in your guest OS, it will crash). A simple shutdown then start immediately after will make it boot with all the currently available CPU instructions/features

@jag5cof said in disk failure: @DustinB I have only one raid, (software) Truenas scales. What is XCP-ng installed onto? I'm not particularly familiar with the Truenas Scales platform, you may be better to reach out to their team/community for guidance. But generally speaking, two failed drives in an array, means you've already lost data. Your best bet would be to rebuild the array and then restore your data from backup.

@olivierlambert thank you

Ah I see, so if you enable it in the BIOS but without loading the module in the Dom0, then it will think the dead man switch is triggered and you are in for a reboot, I see Thanks for sharing your experience anyway!

@Niko1 Glad you're having a good experience with it! I have more recently done GPU passthrough and it's been fantastic actually so I take back a bit of what I had said before. For some reason in the past I was having major issues with it and I think I may have had defective hardware (very hard to rule out in this situation though).