@olivierlambert I am asking XO Team about this issue.

@chicagomed said in Passthrough Contention Problems with Console and Linux VM: @TeddyAstie great will take a look this weekend. Is there anything in particular you want us to test / check out? What works/doesn't work and overall performance.

@jimmymiller i just noticed one of your pics of the backup had NDB enabled. (says exporting content of VDI shifter through NBD. Can you check in your backup job if NDB is on? [image: 1755550405681-f90df5d8-81e6-4404-ba28-c337c0702aad-image.png] I would turn it on and off a few times and leave it off then save the job and retry. There was issues with NDB gui in prior version that might be still affecting your test backup job.

@yann said in New Rust Xen guest tools: @john.c OK, that will be useful when the repo is signed, but for now I don't see what adverse effect it can have. Do I miss something? Also we try to avoid breaking support for older OS versions, so we'll likely continue to advertise the old format for older versions of Debian. @yann From Debian 13.0.0 (code name Trixie) having repository signing is mandatory. Without it apt will straight refuse to install, update or upgrade its packages. Also doing with deb822 format will help to protect the GPG Key, used by Vates from abuse by another repository. Especially if that repository is hosting malware laden deb packages. As only the Vates repository can then use that signing key, as defined in the sources file. Refusing to install, update or upgrade is an adverse effect wouldn’t you say?

That's nice! It means it scales relatively well with many threads, it's a good result

@AtaxyaNetwork I appreciate the sentiment. I think this one is all on me as pointed out by @Danp .. My VDI's would not register without this step. I'm unsure as to why because the error logs were completely blank within XO. Your post in conjunction with the docs were extremely helpful though!

@Bastien-Nollet Thanks as always @florent

Sometimes it when asking questions as well as hallucinating ChatGPT, can respond with results based on old versions (not updated for an up to date code base). So if any past in development or experimental code branch, had its code and was mistakenly released. Then the mistake was found and fixed, it can have found its way into the training data set. Can also be an example of training data set poisoning, forcing it to give misleading or mistaken responses, due to it hallucinating as a result.

Ok just to close the loop here. Problem persisted with Intel Arc (the only thing I forgot to do was disable REBAR, but I don't think this is a related issue). Upon installing Radeon RX 6400, pass through worked without any issue. Of note, on the Arc I got two separately numbered PCI devices 0000:0b and 0000:0a where as with the AMD it was 0000.08.01 and .02. I'm not sure if that helps hint at the issue --something to do with grouping and not grouping? This is way above my head but I noticed the difference. I'm still having issues with pass through into Linux but that is a separate topic. Looks like we have a good discussion of this issue on the other post and so I will stay on that one going forward. Thank you!

@MK.ultra don't think so it's working without for me.

Excellent news!

@olivierlambert While VDI is maybe not as vital as it once was...I'm experimenting with multimedia work in XCP-ng. Having a VM with GPU off-loading of CODEC encoding would be nice. It's a pretty big CPU hit to make that go.

@DustinB Yeah I thought about doing something like this, but then we are getting into overkill costs lol. My main point behind this post was to get info on what other people are doing, I presumed most are just relying on config backups and will restore it at the other site if needed, which is fine. I was just also debating about whether or not I wanted the subnets to match between these 2 sites, since that can create some routing headaches, but should still be doable and would make recovery even faster. @olivierlambert yeah I do use that feature, so that's very helpful! Should make this all doable.

@Danp Hi, I'll try this. Thank you very much. Best regards Mattmurdock78

@Danp Thank you.

No luck, I've tried building from source, using the script, etc. Following along with Google Gemini I even managed to get walked into the process of damaging the Xapi to the point of losing access to the server from Xen Orchestra / XCP-ng Center (unfortunately for me it was the Pool Master). Thankfully I had backups on my NAS and experienced minimal loses through the process of having to reinstall the operating system. At this point, I'm guessing my pool that dates back to version 8.1 is screwed up in some way beyond human or AI understanding. I have 2 Servers & 3 NAS units (2 TrueNas Scale) + 1 Synology that all work with the old version of Xen Orchestra but I can't get it to work with anything else. Manually putting the same info in or importing the configuration. One of the servers was rebuilt today and added back to the pool. Is it possible that reinstalling the OS on the other server could fix everything? I'm confused because the backups aren't failing from the old Xen Orchestra. Old: [image: 1756784257357-19e6f3da-3b22-4f7d-a47e-66d8216b3f8b-image.png] New: [image: 1756784292273-263493de-a5cb-4610-b7ba-ba173fe7b055-image.png] Any help and / or ideas would be greatly appreciated. Thanks!

@olivierlambert Great to know it's not something I was doing wrong! Hopefully these other cards I have ordered will afford me more luck! Cheers

Thank you everyone for your feedback, really appreciated!

@olivierlambert Excellent news!!! I will keep my eyes on that, and provide a tests! Thank you!

That sounds interesting, the backups are on a new remote on a new Minio-instance so the issue is handled for now. I guess we will see over time if the backup times increases.