Vr2Io

Power only, won't use PCIe.

Quite affordable then what I expect. "Highpoint" come back again for interesting stuff.

All lane is 12gb, i.e. 8i card means support 8x 12gb max. All depends on the device speed i.e. 6 or 12gb. For 6gb, single cable also provide over 1.5GB/s. In simple say, 12gb only useful between controller and expander when both is 12gb.

No issue, if all connect well. All port / lane were 12gb. No big difference, 12gb in general only useful for 12gb expander, because device often 6gb and haven't use up that bandwidth. 6gb single cable have ~2GB/s throughput, so slightly bottleneck for 20 disks. You can go to dual cable or 12gb, actually depends which you like.

I already said use it to power up the expander in external storage enclosure. Nothing relate the PCIe signal and USB cable. If you have electronic skill, you can connect the PCIe to PSU without those adaptor. In fact, expander were best solution to connect many SAS/SATA device, I haven't got slowing.

Expander won't use PCIe signal, you just need power it up. i.e. PCIe riser card, no need USB part.

Fix this first.

You post a SMART info. but none have 187 and 198 was "0".

Thats true. Suppose problem cause by WD SATA-USB bridge, other bridge should be fine. I shuck lot of WD USB, but not on last 4 because local sale for raw disk even cheaper. Also if you rebuild it under WD USB bridge, then you can't shuck the disk and simple put back to array, it will unmountable.

Once breakthrough happen, it got interesting. I setup 12x 3.5" disk RAID0, it also up to ~2.1GB/s.

Disk look like fine by its SMART info., but have drop at 01:57 1-Jan Jan 1 01:58:02 unKevin unassigned.devices: Disk with ID 'ST8000AS0002-1NA17Z_Z840FGMZ (sdq)' is not set to auto mount. Jan 1 07:05:55 unKevin emhttpd: error: hotplug_devices, 1719: No such file or directory (2): Error: tagged device ST8000AS0002-1NA17Z_Z840FGMZ was (sdj) is now (sdq)

Dec 31 17:22:15 Tower kernel: sd 0:0:0:0: [sda] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=12s Diagnostic have 14TB disk SMART and indicate fine. According above message, it look like some timeout issue happen. Anyway if endup not use USB ( JMicron controller ), then suggest connect it there and preclear ( if you consider necessary ).

You should update the disk firmware

Usually are bad drive, you should check the SMART. But Unraid usually can't read thr USB-SATA bridge, so best should put to Windows and check the SMART.

As OP found stock memtest86 not very well to exam have error or not, it have newer version more aggressive. I also found same result by cross check with other memory test program. Very quickly detect error in less then 2hr.

Out of memory occur, some process have crash, you should reboot first. Dec 30 08:28:46 FreenasMedia nginx: 2022/12/30 08:28:46 [crit] 10382#10382: ngx_slab_alloc() failed: no memory Dec 30 08:28:46 FreenasMedia nginx: 2022/12/30 08:28:46 [error] 10382#10382: shpool alloc failed Dec 30 08:28:46 FreenasMedia nginx: 2022/12/30 08:28:46 [error] 10382#10382: nchan: Out of shared memory while allocating message of size 522. Increase nchan_max_reserved_memory. Dec 30 08:28:46 FreenasMedia nginx: 2022/12/30 08:28:46 [error] 10382#10382: *208992 nchan: error publishing message (HTTP status code 500), client: unix:, server: , request: "POST /pub/cpuload?buffer_length=1 HTTP/1.1", host: "localhost"

Yes all lost, in fact no matter which storage technic, it can crash/corrupt, so backup was important. Raid0 just for speed, on store and recovery. Btrfs RAID0 no need all disk in same size.

Raid 5/6 also support in btrfs. But I don't like because its performance not great and need sync/rebuild in initial. I haven't use any cache pool / mover. Or you can wait Unraid 6.12, seems officially support ZFS, I may try that in future.

Suggest only use two disk, 1 parity and 1 data. Then actually they just a pair mirror disk. 8TB disk also in reasonable price. It could be SMR.

1g NIC can reach 100MB+/s transfer speed 10g NIC can reach 1GB+/s transfer speed You can create raid0 ( or other raid mode ) btrfs disk pool then mount it by UD, but any disk fail will lost all data in that pool.

No problem for 9220-8i or 9240-8i, but we usually flash it to IT mode. Passthrough to either one or you need two HBA. 7x24 ? Otherwise, not much power save.

Yes, I mixed-up the name SVM.

Try disable SVM (IOMMU) then check any different. If positive, then re-try all option in SVM.

Right, press some hot key sequence may continue booting.

This need BIOS pcie bifurcation support. Or you may try M2 to PCIe slot Or USB display device.