I am currently using a Lenovo ThinkStation P920 for my home lab setup, but I’m facing a limitation with the number of drives I can install.
The Qnap TR-004 has caught my attention because it can be configured as a JBOD, allowing me to add four new 12TB hard drives. My plan is to connect the Qnap TR-004 to my P920, where I have Proxmox 8.3.0 installed, and pass the drives from the TR-004 through to a VM running TrueNAS.
Challenges other users are Facing:
After rebooting the entire Proxmox system, the VM, or the TR-004, the setup stops working as it did before. Can anyone explain why this happens and how to resolve it?
I’ve read that passing through the hard disks can reduce performance. Is this true, and if so, to what extent?
Main Question: Overall, will this setup work effectively with good to moderate performance? Are there any potential impacts on the hard drives I plan to install in the TR-004?
Thank you to everyone who takes the time to read this and provide helpful comments.
Isn’t the TR-004 known to only expose a single serial number even though there are more drives in the enclosure? That in turn means TrueNAS can’t adress the drives properly.
I stay away from multi-drive USB-enclosures because between the serial number issue and it simply being USB they don’t appear to be reliable enough, at least not to the level I want. I stick to connecting my drives to motherboard SATA or when I run out of those, to dedicated LSI-controllers (3008-based ones being my preferred type).
While I can’t give you any more pointers you might be able to google some success stories, I’d be surprised if someone hasn’t looked into it and shared their experiences somewhere.
I happen to have a left-over QNAP TR-004 in my homelab as well.
It’s configured as “individual disks” (all jumpers to the left). When I connect it to TrueNAS SCALE 24.10.2, the GUI counts “2” unused disks (correct), but only one is shown under Storage/Disks, and none is offered for pool creation.
It gets more interesting at shell level:
$ sudo lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS
..
sdf 8:80 0 9.1T 0 disk
└─sdf1 8:81 0 9.1T 0 part
sdg 8:96 0 9.1T 0 disk
└─sdg1 8:97 0 9.1T 0 part
..
$ sudo smartctl -d jmb39x-q,2 --all /dev/sdf
=== START OF INFORMATION SECTION ===
Model Family: Western Digital Red
Device Model: WDC WD100EFAX-68LHPN0
Serial Number: 2YJ8B1UD
...
$ sudo smartctl -d jmb39x-q,3 --all /dev/sdg
=== START OF INFORMATION SECTION ===
Model Family: Western Digital Red
Device Model: WDC WD100EFAX-68LHPN0
Serial Number: 2YJ95ZJD
So, both disks are correctly recognized by the system, including their serial numbers.
Given that I cannot use them for pool creation, is there another way to use them? Can I pass something like /dev/sdg to a jail? Or maybe to an incus vm, once those are available?