Hi Guys,
discovered Unraid a few weeks ago and thought I will give it a shot, so that I can probably get rid of my power hungry HP Server with ESXi running my VMs. As the HP has hardware raid (no JBOD option), I decided to buy new hardware to test Unraid in parallel.
I bought:
be quiet! Pure Power 11 400W (80Plus Gold)
AORUS B550 AORUS ELITE V2 (board has 2x M.2 and 4x SATA, not shared in any way)
AMD Ryzen™ 5 3600 (6c/12t)
MSI GeForce GT 710 1GD3H LP
32GB ECC RAM (2x Mushkin DIMM 16 GB DDR4-2133 ECC)
M.2 NVMe 1TB (Apacer AS2280P4 1 TB, PCIe 3.0 x4, NVMe 1.3, M.2 2280) --> M.2 Slot M2A-CPU
M.2 NVMe 1TB (Apacer AS2280P4 1 TB, PCIe 3.0 x4, NVMe 1.3, M.2 2280) --> M.2 Slot M2B_SB
Seagate BarraCuda 4 TB ST4000DM004 (SATA 6 Gb/s, 3,5") --> SATA 0
Seagate BarraCuda 4 TB ST4000DM004 (SATA 6 Gb/s, 3,5") --> SATA 1
GIGABYTE SSD 240 GB (SATA 6 Gb/s, 2,5") --> SATA 2
Setup in Unraid (v6.9.2)
Array: 1x 4TB HDD
Parity: 1x 4 TB HDD
Cache Pool: 1 TB NVMe + 1 TB NVMe
Unassigned: 240GB SATA SSD
Besides creating the Parity Disk (took a few hours) and formatting the Array, I have not done anything yet, besides purchasing a license 😄
The problem that I currently have, is that one of the NVMe drives is suddenly missing from the cache 🤯
After a reboot I see both drives there again, few minutes later, again missing.
The log is showing this:
Jul 2 08:47:53 Unraid kernel: blk_update_request: I/O error, dev nvme0n1, sector 2000409088 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 0
Jul 2 08:47:53 Unraid kernel: nvme 0000:01:00.0: can't change power state from D3hot to D0 (config space inaccessible)
Jul 2 08:47:53 Unraid kernel: nvme nvme0: Removing after probe failure status: -19
Jul 2 08:47:53 Unraid kernel: Buffer I/O error on dev nvme0n1, logical block 250051136, async page read
Jul 2 08:47:53 Unraid emhttpd: error: ckmbr, 2197: Success (0): read: /dev/nvme0n1
Jul 2 08:47:53 Unraid emhttpd: shcmd (103): mkdir -p /mnt/cache
Jul 2 08:47:53 Unraid kernel: nvme nvme0: failed to set APST feature (-19)
Jul 2 08:47:53 Unraid emhttpd: /mnt/cache mount error: No pool uuid
This is during bootup before:
Jul 2 08:40:31 Unraid kernel: nvme nvme0: pci function 0000:01:00.0
Jul 2 08:40:31 Unraid kernel: scsi host2: ahci
Jul 2 08:40:31 Unraid kernel: nvme nvme1: pci function 0000:04:00.0
Jul 2 08:40:31 Unraid kernel: RAPL PMU: API unit is 2^-32 Joules, 1 fixed counters, 163840 ms ovfl timer
Jul 2 08:40:31 Unraid kernel: scsi host3: ahci
Jul 2 08:40:31 Unraid kernel: scsi host4: ahci
Jul 2 08:40:31 Unraid kernel: RAPL PMU: hw unit of domain package 2^-16 Joules
Jul 2 08:40:31 Unraid kernel: scsi host5: ahci
Jul 2 08:40:31 Unraid kernel: scsi host6: ahci
Jul 2 08:40:31 Unraid kernel: ata1: SATA max UDMA/133 abar m131072@0xfc780000 port 0xfc780100 irq 48
Jul 2 08:40:31 Unraid kernel: ata2: SATA max UDMA/133 abar m131072@0xfc780000 port 0xfc780180 irq 48
Jul 2 08:40:31 Unraid kernel: ata3: SATA max UDMA/133 abar m131072@0xfc780000 port 0xfc780200 irq 48
Jul 2 08:40:31 Unraid kernel: ata4: SATA max UDMA/133 abar m131072@0xfc780000 port 0xfc780280 irq 48
Jul 2 08:40:31 Unraid kernel: ata5: SATA max UDMA/133 abar m131072@0xfc780000 port 0xfc780300 irq 48
Jul 2 08:40:31 Unraid kernel: ata6: SATA max UDMA/133 abar m131072@0xfc780000 port 0xfc780380 irq 48
Jul 2 08:40:31 Unraid kernel: ccp 0000:08:00.1: ccp: unable to access the device: you might be running a broken BIOS.
Jul 2 08:40:31 Unraid kernel: ACPI Warning: SystemIO range 0x0000000000000B00-0x0000000000000B08 conflicts with OpRegion 0x0000000000000B00-0x0000000000000B0F (\GSA1.SMBI) (20200925/utaddress-204)
Jul 2 08:40:31 Unraid kernel: ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
Jul 2 08:40:31 Unraid kernel: libphy: r8169: probed
Jul 2 08:40:31 Unraid kernel: r8169 0000:05:00.0 eth0: RTL8125B, 18:c0:4d:ad:ea:b6, XID 641, IRQ 55
Jul 2 08:40:31 Unraid kernel: r8169 0000:05:00.0 eth0: jumbo features [frames: 9194 bytes, tx checksumming: ko]
Jul 2 08:40:31 Unraid kernel: nvme nvme0: missing or invalid SUBNQN field.
Jul 2 08:40:31 Unraid kernel: nvme nvme1: missing or invalid SUBNQN field.
Jul 2 08:40:31 Unraid kernel: cryptd: max_cpu_qlen set to 1000
Jul 2 08:40:31 Unraid kernel: r8125: loading out-of-tree module taints kernel.
Jul 2 08:40:31 Unraid kernel: AVX2 version of gcm_enc/dec engaged.
Jul 2 08:40:31 Unraid kernel: AES CTR mode by8 optimization enabled
Jul 2 08:40:31 Unraid kernel: nvme nvme0: allocated 64 MiB host memory buffer.
Jul 2 08:40:31 Unraid kernel: nvme nvme1: allocated 64 MiB host memory buffer.
Jul 2 08:40:31 Unraid kernel: nvme nvme0: 8/0/0 default/read/poll queues
Jul 2 08:40:31 Unraid kernel: nvme nvme1: 8/0/0 default/read/poll queues
As everything is brand-new, those drives are not tested in other scenarios (like Windows install), yet.
Of course it could be a faulty device, but it's probably not soo likely.
Has anybody seen this behavior before?
Cheers,
Jack
Overview of Main tab:
BIOS is latest version, for the NVMe it seems that there is no firmware update available.
BIOS detected both drives: