На ядре линукс 6.6.11 команда fstrim -v -a даёт неожиданный результат
[ 361.729520] nvme nvme0: I/O 768 (Dataset Management) QID 6 timeout, aborting
[ 361.729542] nvme nvme0: I/O 769 (Dataset Management) QID 6 timeout, aborting
[ 361.729549] nvme nvme0: I/O 770 (Dataset Management) QID 6 timeout, aborting
[ 361.729556] nvme nvme0: I/O 771 (Dataset Management) QID 6 timeout, aborting
[ 361.729709] nvme nvme0: Abort status: 0x0
[ 361.729728] nvme nvme0: Abort status: 0x0
[ 361.729733] nvme nvme0: Abort status: 0x0
[ 361.729754] nvme nvme0: Abort status: 0x0
[ 361.821488] nvme nvme0: I/O 799 (Dataset Management) QID 6 timeout, aborting
[ 361.821501] nvme nvme0: I/O 800 (Dataset Management) QID 6 timeout, aborting
[ 361.821507] nvme nvme0: I/O 801 (Dataset Management) QID 6 timeout, aborting
[ 361.821512] nvme nvme0: I/O 802 (Dataset Management) QID 6 timeout, aborting
[ 361.821685] nvme nvme0: Abort status: 0x0
[ 361.821704] nvme nvme0: Abort status: 0x0
[ 361.821708] nvme nvme0: Abort status: 0x0
[ 361.821727] nvme nvme0: Abort status: 0x0
[ 391.934516] nvme nvme0: I/O 768 QID 6 timeout, reset controller
[ 399.661516] pcieport 0000:00:01.1: AER: Uncorrected (Fatal) error received: 0000:00:01.1
[ 399.661532] pcieport 0000:00:01.1: PCIe Bus Error: severity=Uncorrected (Fatal), type=Transaction Layer, (Receiver ID)
[ 399.661537] pcieport 0000:00:01.1: device [8086:2f03] error status/mask=00000020/00000000
[ 399.661543] pcieport 0000:00:01.1: [ 5] SDES (First)
[ 399.661551] nvme nvme0: frozen state error detected, reset controller
[ 455.937017] nvme0n1: Read(0x2) @ LBA 2350616, 8 blocks, Host Aborted Command (sct 0x3 / sc 0x71)
[ 455.937032] I/O error, dev nvme0n1, sector 2350616 op 0x0:(READ) flags 0x81700 phys_seg 1 prio class 2
[ 455.943572] nvme 0000:02:00.0: Unable to change power state from D3cold to D0, device inaccessible
[ 455.943694] nvme nvme0: Disabling device after reset failure: -19
[ 455.949596] I/O error, dev nvme0n1, sector 501375865 op 0x1:(WRITE) flags 0x29800 phys_seg 1 prio class 2
[ 455.949615] XFS (nvme0n1p3): log I/O error -5
[ 455.949622] XFS (nvme0n1p3): Filesystem has been shut down due to log error (0x2).
[ 455.949626] XFS (nvme0n1p3): Please unmount the filesystem and rectify the problem(s).
[ 456.977574] pcieport 0000:00:01.1: broken device, retraining non-functional downstream link at 2.5GT/s
[ 457.010546] pcieport 0000:00:01.1: Data Link Layer Link Active not set in 1000 msec
[ 457.010557] pcieport 0000:00:01.1: AER: Root Port link has been reset (-25)
[ 457.010565] pcieport 0000:00:01.1: AER: subordinate device reset failed
[ 457.010593] pcieport 0000:00:01.1: AER: device recovery failed
В результате файловые системы на отвалившихся дисках становятся недоступны со всеми вытекающими.
На ядрах 6.5.Х всё работает. Продолжаю наблюдения.
