Search results

I recently bought a trio of Dell Optiplexes that have 10th gen CPUs in them and got mediated passthrough to work by following this tutorial: https://3os.org/infrastructure/proxmox/gpu-passthrough/igpu-split-passthrough/ Note the two comments I made at the bottom of the page

Nope. I gave up.

The pool field can be the path to any *dataset*, not a ZVOL. The plugin creates the zvols in the specified pool and manages the iscsi stuff.

Somewhat related but I didn't know where else to ask this: Where can I make feature requests?

I believe that, I just have no idea what needs tuning or what the bottleneck even is.

Yeah any operations over iSCSI was quite slow for me, even over a 10g link. NFS is much faster.

I found this: https://sourceforge.net/p/scst/mailman/message/35242474/ I'm not sure if it's relevant as I don't really understand what Vladislav B is saying but from what I can guess qemu is asking the iscsi target about the status of the block at LBA 0 and the target responds...

I wouldn't call that ZFS over iSCSI because ZFS isn't being overlaid iSCSI. In any case I did manage to get it working. I'll have to make a write-up on how to get this mess (no offence intended to surfrock66) working but the thing that I got wrong was creating an extent, you don't need to do...

Am I supposed to create a dataset or a ZVOL? I tried a ZVOL and got the "parent is not a filesystem" error (yeah, the whole point of ZFS over iSCSI is that ZFS is put on top of the iSCSI block device) and I tried a dataset but a dataset was no good. There's missing information here.

My switch does support jumbo frames, I just can't find the setting for it in the Omada controller.

I've got a test machine with a 10 gig NIC and two 1 gig NICs. The 10 gig NIC is connected to a 10 gig interface on my switch and one of the gigabit NICs is also connected to the same switch. As part of my testing I have set up these two NICs in an active-backup bond like so: auto lo...

The actual boot disk (scsi0) is about 2.2GB in size but I want to make it bigger so that I can store stuff on it like kubernetes.

Okay I don't know what exactly it was from your initial command that got it working for me but it's working now: wget https://cloud-images.ubuntu.com/jammy/current/jammy-server-cloudimg-amd64.img qm create 8000 --memory 2048 --balloon 0 --core 2 --bios ovmf --machine q35 --cpu host --numa 1...

I'm trying to automate the creation of VMs using cloudimg's and I've got a template and a clone created, but the clone VM won't start it just says "starting serial terminal on interface serial0". This is my script: wget...

What a weird setting, surely you wouldn't want your windows VMs to crash

What's considered an older CPU? Also out of curiosity what does that setting do?

Now that I think about it I had vanilla windows server 2019 VMs crash on me too but I don't know if those crashes were caused by the same thing. I'll spin up a couple and see if any of them crash. [Edit] I tried starting up my two domain controller testing VM's that are both running server 2019...

The VM is running Server 2022 so yes UEFI with pre-enrolled keys.

I think I'm having this issue with my poweredge R730XD which has dual E5-2640 v3's. I get the same error as the OP about the "hardware error 0x80000021". I am also using 5.15.35-1-pve.

I've noticed that the NoVNC console works just fine for windows VMs, but not Linux VMs. No idea about BSD VMs.