Recent content by Domino

The nag doesn't hurt me per-se, but yes I have a fair few installations across my lab. If one off payment per year covers my whole lab that would be pretty awesome! and nobody, even the slimmest of wallets could bat an eye lid at offering over some cash every year. :cool: Kind of like no...

Where it doesn't have any connection to financials as such, something like a $30 dollar per year thing? then at least we can get rid of the pop up message that notifies it is not a subscription/licensed install (of course one could work around that, but i digress), PLUS most importantly It would...

As the first node has been running fine, I now updated another lab node, this time it is an AMD 5950X in an Asus WS-x570-pro-ace board. I noted a similar trend where the first reboot after upgrade was a little slower than usual, but subsequent reboots were fine. In this setup all the resource...

Nice! It's been years since I last used BTRFS with anger, will give your script a run in my lab, thanks! We need to see more instructional step-by-step writeups on here, more resources the better.

I had a similar issue with a usb modem a few years back, in the end I resorted to passing through a whole usb controller via vfio passthrough, luckily the board had 2 separate iommu groups for the controllers, so the host kept one and guest got the other, problem was solved in that particular...

Exactly the kind of gear I like to lab too, however I've been stung a few times jumping onto a 'newish' whitebox model board when it comes to Linux. Oddly enough I've never had a single issue with Windows server, I wonder how big a share does Microsoft still have in the server enterprise world.....

Looks like its a little bit further now, devils advocate, did you try the memtest? Now that you have an installed OS, other things to try... going into bios and disabling anything relative to apic, ie x2apic etc., disable iommu and all virtualization options, also disable hyperthreading...

Stick to the default 'io_uring', it is non-blocking these days and is optimal for virtio-scsi-single, in some heavy load cases (overprovisioned), native can help somewhat, otherwise io_uring is your best option. Also I tend to change the cache to Direct-Sync for boot-disks for peace of mind...

Another way you could try..... Load up a proxmox or hyperv or virtualbox on another computer, and either passthrough the m.2 boot drive or later on just re-image the m.2 boot drive from the virtual drive via clonezilla etc. So basically image the drive with proxmox preinstalled, but remember...

Went from 6.14 (6.14.11-3-pve) to the 6.17 (6.17.2-1-pve) in the pve repository as of right now. Hardware: Intel Platform (Vendor: HP) CPU: i5 13500T Did a dmesg comparison, discovered these 'differences': 1) In 6.17 I dont see the line 'zbud: loaded' - which I believe has something to do...

@mikeyo , not sure if you have considered it, but in my case I configured a standalone data tier server cluster, from there I present block devices via nvmeof-rdma/iscsi/smb to my compute servers across a 400G backbone. The only difficult thing was that I just couldn't find a software system...

The problem with Microsoft Windows is that I do not believe it is in their list of priorities to make sure they don't break memory ballooning for QEMU/KVM across their update cycles, for them its all about Hyper-V (their own virtualisation layer). So basically any 'ballooning' explosion...

If you want the 'best' performance from the drives, ZFS is not going to cut it. ZFS is about ultimate data resiliency, it still to this day has not been completely optimized for NVME drives. I would suggest creating a ZFS pool for vm bootdisks, where you just use SAS/SATA SSDs as main storage...

As far back as I remember, memory ballooning has always been a little 'buggy' with windows, the general consensus is to leave it disabled for Windows guests.

Beautiful modern UI!!!! I don't run a datacenter, just a homelab, however this makes it easier to keep an eye on things across all nodes from a central dashboard. Some suggestions if I may: 1) could the Resources section of the remote-view have sub-groups under the node rather than list all...