CommentStreams:68d980a9553cbd7667b7858fcab2251e

I have tried to configure this X1011 board with a RPi5B within the C1 case for nearly 2 weeks now - roughly 60 hours of testing. The NVMe SSDs function just fine as independent drives, but I have not been able to write to two or more SSDs simultaneously, not matter my configuration settings. I've tried to do so using three different sets of four SSDs: Fikwot 1TB SSDs, MMoment 1TB SSDs, and Goldenfir 256GB SSDs. In fact, I also could not create any filesystem for a software RAID0 across all four SSDs. The efforts error out with errors like this most ofttimes:

[ 0.517966] nvme nvme0: pci function 0000:03:00.0 [ 0.517973] nvme 0000:03:00.0: enabling device (0000 -> 0002) [ 0.521768] nvme nvme0: 3/0/0 default/read/poll queues [ 0.522660] nvme nvme0: Ignoring bogus Namespace Identifiers [ 0.523600] nvme0n1: p1 p2 [ 0.523916] nvme nvme1: pci function 0000:04:00.0 [ 0.523924] nvme 0000:04:00.0: enabling device (0000 -> 0002) [ 0.527246] nvme nvme1: 1/0/0 default/read/poll queues [ 0.527722] nvme nvme1: Ignoring bogus Namespace Identifiers [ 0.533278] nvme1n1: p1 p2 [ 0.533647] nvme nvme2: pci function 0000:05:00.0 [ 0.533657] nvme 0000:05:00.0: enabling device (0000 -> 0002) [ 0.537068] nvme nvme2: 1/0/0 default/read/poll queues [ 0.537614] nvme nvme2: Ignoring bogus Namespace Identifiers [ 0.541164] nvme2n1: p1 p2 [ 0.541496] nvme nvme3: pci function 0000:06:00.0 [ 0.541504] nvme 0000:06:00.0: enabling device (0000 -> 0002) [ 0.544821] nvme nvme3: 1/0/0 default/read/poll queues [ 0.545306] nvme nvme3: Ignoring bogus Namespace Identifiers [ 0.549162] nvme3n1: p1 p2

[ 439.140011] [<000000009db9d36e>] nvme_irq [ 439.140017] [<000000009db9d36e>] nvme_irq [ 439.140020] [<000000009db9d36e>] nvme_irq [ 439.140023] [<000000009db9d36e>] nvme_irq

[ 470.187812] nvme nvme1: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10 [ 470.187820] nvme nvme1: Does your device have a faulty power saving mode enabled? [ 470.187823] nvme nvme1: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug [ 470.187836] nvme nvme3: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10 [ 470.187839] nvme nvme3: Does your device have a faulty power saving mode enabled? [ 470.187841] nvme nvme3: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug [ 470.187854] nvme nvme2: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10 [ 470.187857] nvme nvme2: Does your device have a faulty power saving mode enabled? [ 470.187858] nvme nvme2: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug [ 470.263828] nvme 0000:05:00.0: enabling device (0000 -> 0002) [ 470.266477] nvme nvme2: 1/0/0 default/read/poll queues [ 470.271909] nvme 0000:06:00.0: enabling device (0000 -> 0002) [ 470.272349] nvme 0000:04:00.0: enabling device (0000 -> 0002) [ 470.273677] nvme nvme2: Ignoring bogus Namespace Identifiers [ 470.275687] nvme nvme1: 1/0/0 default/read/poll queues [ 470.276323] nvme nvme3: 1/0/0 default/read/poll queues [ 470.284265] nvme nvme1: Ignoring bogus Namespace Identifiers [ 470.291824] nvme nvme3: Ignoring bogus Namespace Identifiers [ 501.248802] nvme nvme0: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0xffff [ 501.248809] nvme nvme0: Does your device have a faulty power saving mode enabled? [ 501.248811] nvme nvme0: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug [ 501.248811] nvme nvme2: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10 [ 501.248815] nvme nvme2: Does your device have a faulty power saving mode enabled? [ 501.248817] nvme nvme2: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug [ 501.538195] nvme nvme3: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10 [ 501.538197] nvme nvme3: Does your device have a faulty power saving mode enabled? [ 501.538198] nvme nvme3: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug [ 501.538213] nvme nvme1: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10 [ 501.538215] nvme nvme1: Does your device have a faulty power saving mode enabled? [ 501.538217] nvme nvme1: Try "nvme_core.default_ps_max_latency_us=0 pcie_aspm=off pcie_port_pm=off" and report a bug [ 501.887998] nvme 0000:06:00.0: enabling device (0000 -> 0002) [ 501.888234] nvme 0000:04:00.0: enabling device (0000 -> 0002) [ 501.888428] nvme 0000:05:00.0: enabling device (0000 -> 0002) [ 501.890650] nvme nvme2: 1/0/0 default/read/poll queues [ 501.890963] nvme nvme1: 1/0/0 default/read/poll queues [ 501.892727] nvme nvme1: Ignoring bogus Namespace Identifiers [ 501.893103] nvme nvme3: 1/0/0 default/read/poll queues [ 502.204741] nvme nvme2: Ignoring bogus Namespace Identifiers [ 503.362417] nvme nvme3: Ignoring bogus Namespace Identifiers [ 504.582800] nvme 0000:03:00.0: Unable to change power state from D3cold to D0, device inaccessible [ 506.897825] nvme nvme0: Disabling device after reset failure: -19 [ 506.924122] I/O error, dev nvme0n1, sector 383009792 op 0x1:(WRITE) flags 0x800 phys_seg 1 prio class 2

These errors and others akin to these happen no matter the ARGs sumbitted to the config.txt and cmdline.txt files. I've attempted to use suggestions on GEEKWORK wiki/forum, and many on usual Linux troubleshooting sites. Even those suggested here within the errors do not improve nor suggest they're making promising progress. These are scripts to write to the four SSDs simultaneously --- cut and paste one line each to separate terminal windows and execute. Each will attempt to write 931GB to each 1TB SSD (each 1TB SSD actually holds 953.86946868896484375GB).

clear; echo; time taskset -c 0 nice dd if=/dev/urandom of=/dev/nvme0n1 bs=2097152 count=487936 conv=notrunc status=progress & clear; echo; time taskset -c 1 nice dd if=/dev/urandom of=/dev/nvme1n1 bs=2097152 count=487936 conv=notrunc status=progress & clear; echo; time taskset -c 2 nice dd if=/dev/urandom of=/dev/nvme2n1 bs=2097152 count=487936 conv=notrunc status=progress & clear; echo; time taskset -c 3 nice dd if=/dev/urandom of=/dev/nvme3n1 bs=2097152 count=487936 conv=notrunc status=progress &

Create a software RAID0 across created partitions on all four SSDs:

mdadm --create --verbose /dev/md0 --level=0 --raid-devices=4 /dev/nvme0n1p1 /dev/nvme1n1p1 /dev/nvme2n1p1 /dev/nvme3n1p1

Then attempt to create a filesystem:

mkfs.ext2 /dev/md0p1

Or after a reboot:

mkfs.ext2 /dev/md127p1

AND... The errors persist.

I have had no success attempting to do these things.