nvme2 broken on tb-build-02

it's running with a degraded array now:

[1482255.553273] nvme nvme2: I/O tag 769 (5301) opcode 0x9 (I/O Cmd) QID 30 timeout, aborting req_op:DISCARD(3) size:67108864
[1482255.573963] nvme nvme2: I/O tag 770 (e302) opcode 0x9 (I/O Cmd) QID 30 timeout, aborting req_op:DISCARD(3) size:17825792
[1482287.309436] nvme nvme2: I/O tag 769 (5301) opcode 0x9 (I/O Cmd) QID 30 timeout, reset controller
[1482408.717900] nvme nvme2: Device not ready; aborting reset, CSTS=0x1
[1482408.777946] nvme nvme2: Abort status: 0x371
[1482408.783499] nvme nvme2: Abort status: 0x371
[1482468.794187] nvme nvme2: Device not ready; aborting reset, CSTS=0x1
[1482470.026274] nvme 0000:81:00.0: not ready 1023ms after FLR; waiting
[1482471.082341] nvme 0000:81:00.0: not ready 2047ms after FLR; waiting
[1482473.386530] nvme 0000:81:00.0: not ready 4095ms after FLR; waiting
[1482477.738284] nvme 0000:81:00.0: not ready 8191ms after FLR; waiting
[1482486.186306] nvme 0000:81:00.0: not ready 16383ms after FLR; waiting
[1482496.196099]  nvme_hwmon_read+0x58/0x1a0 [nvme_core]
[1482496.219572]  nvme_hwmon_read+0x58/0x1a0 [nvme_core]
[1482496.243081]  nvme_hwmon_read+0x58/0x1a0 [nvme_core]
[1482504.362516] nvme 0000:81:00.0: not ready 32767ms after FLR; waiting
[1482539.178674] nvme 0000:81:00.0: not ready 65535ms after FLR; giving up
[1482545.183425] nvme nvme2: Disabling device after reset failure: -25
[1482545.227116] md/raid1:md127: nvme2n1p1: rescheduling sector 44048
[1482545.234893] md/raid1:md127: redirecting sector 44048 to other mirror: nvme3n1p1
[1482545.240517] md/raid1:md127: Disk failure on nvme2n1p1, disabling device.