NVME Controller is down on 4096b sector size in linux!

Hello everyone

I am using a WD BLUE SN 580 nvme…

I recently formatted the nvme with lbaf=1 using nvme cli but when I started writing data to it, it showed some error in dmesg and the nvme became inaccessible …

Reverting to 512b sector size works fine but the error occurs again after changing to 4096b size…

Its also using latest firmware

The error:
[28922.642400] nvme nvme0: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10
[28922.714515] nvme 0000:01:00.0: enabling device (0000 → 0002)
[28922.715277] nvme nvme0: Removing after probe failure status: -19
And also many btrfs io errors

And then the nvme is inaccessible until reboot

Any fix ?

EDIT : The error typed here is copied from another forum but I got the same error

Hard disks and SSDs both use 4096 byte blocks which is by design.

Nope
The sn 580 is using 512b sectors by default and is failing on 4096b sector with the error mentioned above
The nvme controller kept resetting when using 4096b sectors

I have an SN580 and a clean install of Windows uses 4096 sectors

The 512 is emulated and wastes space and has poor performance

It might not be the physical sector size …
I used nvme-cli command in linux to format the drive using 4096b sector size (physical and logical) and every time it threw the same IO error after some time

Read this

Makes me wonder if there is an issue with Linux. I keep a VM with Linux where it is in its own sandbox.

Grab the WD SSD tools for Windows as this SSD has some known issues that firmware update can correct.