@rjt The setting for backup concurrency within the XOA backup settings.

Any chance this could be a network issue? Are you on a flat network or are things segmented and routed between each other?

It may have already been mentioned in here more recently, but I still think some kind of system to select specific VHDs from the ESXi box would be super nice.

Sometimes one may only need to move the data of specific VHDs (excluding say logging or other temporary data style VHDs) and this would greatly speed up migrations.

It would also help in the event someone has VHDs that are larger than 2TiB, since you'd be able to only move the smaller ones and find an alternate solution for the larger than 2TiB VHDs.

Just a thought though; this tool is still amazing and has helped me start migrations!

@brezlord Yes, looks like the disks themselves are thin but whatever was used thick wise is what the disk is inflated to.

i.e. without the thin=true, a 100GB thick disk on ESXi with only 20GB used would end up taking up 100GB on the SR, but if you grow the disk beyond the 100GB it's all thin at that point, so you could change it to a 200GB and it's still using only 100GB of actual space.

BUT with thin=true, the 100GB thick disk on ESXi that only has 20GB used would ONLY use up 20GB of space on the SR in XCP-ng (so it'd only be inflated to 20GB while still showing the OS it's a 100GB disk). Saves a lot of space if you have VMs with massive disks and low usage (like one of the hosts I am migrating which has 13TB assigned thick disks but only about 4TB is used).

@olivierlambert did I get this right?

Just wanna say thanks for the awesome work here, I'm not in a position to test these at the moment but wanted to say thanks nonetheless!

@stormi OK this is a good point. My thing is that I generally don't like to modify things from near default in a prod environment because some update comes along later and breaks it.

To be fair though, I've never had this happen with XCP-ng in my lab and I also doubt something related to certs and secure boot would cause that, so maybe I'm paranoid and incorrect here.

Also, my apologies, I missed the reply about working on a fix.

This all makes sense to me now though, and the fix is pretty easy to get the automated install to work.

Thanks for all the help!

@Darkbeldin Thanks! I followed this response here, which is changing the same line: https://xcp-ng.org/forum/topic/5200/secureboot-certs-install-fails/6

Works fine now!

@andyhhp Will do, I'll link it here once I post it, probably can get that done today once I'm done with work lol.

Thanks for the willingness to help btw!

Hmmmmm I need to look back through some of my old posts and notes but I had a similar issue with a VM with large number of disks. The disks themselves being big wasn't the issue, it was just having a huge number of them that eventually cased a problem. Ended up migrating by moving a single VHD and rebuilding things (the specific system didn't really "need" all the VHDs so it was an easy fix for me, wouldn't be the case here haha).

If you are trying to reduce the number of hosts though, why would migrating the VCSA help? If you're powering off a host then wouldn't you not need the VCSA to be running anyway? Or do you have more than 1 ESXi instance? Might be easiest to just migrate it to another ESXi host if that's the goal, since you won't need the VCSA once you fully move to XCP-ng (assuming you are doing so).

@Andrew This was the answer, thank you! I had it set with 8MiB of video memory which was fine with an older install but 22.04 needs more it looks like. Bumped that up to 16 MiB and it's super smooth now.

Thanks for your help!

@olivierlambert increasing the video memory was the solution. But to answer your question this was in BIOS mode rather than UEFI.

@olivierlambert That's actually something I didn't know about XO Proxy, learn something new everyday haha!

Either way you should have a firewall on both sides right? So you could just use the firewall to whitelist things with rules to the proxy.

I think personally I'd take the VPN route here, but @olivierlambert may disagree and if I'm honest I haven't used XO Proxy much so maybe I'm way off here lol.

Just to be clear, when you try to install the drivers, from AMDs site, for these VMs, it says it can't detect a AMD GPU? Just making sure I know exactly what's going on, not quite sure on a fix but narrowing down the issue will help.

@olivierlambert true, proxies help a lot, I personally still just prefer to avoid things on the WAN as much as I can and IMO VPNs aren't too hard nowadays.

But either is fine for sure.

Also wanted to chime in here that I was seeing about the same speeds when doing some ESXi importing to XCP-ng as well, seems pretty normal to me like @olivierlambert and @florent have already mentioned.

You can do this, first things first though I'd highly recommend using a VPN instead of just letting it do it over the public internet. Wireguard is pretty easy to get going and very fast.

Just wanted to chime in here as I've considered doing this exact thing for years and have never actually managed to be successful enough to go all in.

Not specific to XCP-ng or any issues with it, just GPU passthrough is a huge pain and is likely to create issues down the road so I personally won't do it for my desktop machine as I need that running 100% of the time.

Not trying to put any doubt in here, just that make sure you know what you are doing and know the difficulty that can sometimes occur. I managed to get something similar working on Proxmox once long ago, but it took me literal months of headache and troubleshooting to finally get it functional.

In my mind servers are servers and desktops are desktops, but the idea of putting it all together is still really fun to me so if you do get this working/go through with it, please post about it so we can all hear!

Now spec wise, I agree with others, avoid the 7950X3D for this use case, I have one in my desktop and it's absolutely amazing but I wouldn't use it for a server, may just create some complications (though if you can manage to get it working in a way where the gaming machine/desktop runs on the 3D vCache cores, it'd be a beast).

GPU wise, go with NVidia or AMD IMO, not Intel, there is more information on the 2 other brands when it comes to passthrough, NVidia likes to make it difficult whereas AMD tends to be a bit easier. I've done it with NVidia in the past primarily though.

Maybe also a thought would be to consider a server grade GPU and then use vGPU for this to avoid some of the passthrough headache?

Just wanted to chime in to say thanks for looking into this and "glad" you can reproduce it. If there's anything I can do to help/test let me know, would be happy to.

I've only tested local and SMB so far but both did the same thing.

Looks like this is still happening on the latest XO version, any ideas/updates @olivierlambert ?

@Gheppy you still seeing this as well?

Fix is easy enough for me for now, just rename to .iso and everything is good, just a bit odd.