NVME Read Errors on Parity Check

[Harold_Boom]

Hi all,

 

I have a relatively new nvme drive and I am getting read errors when running my month parity check. Is the drive on the way out?

 

Also should I replace it because I am getting about 1000 parity errors when the parity check runs. System log is full of this;

 

Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009624936
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009624944
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009624952
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009624960
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009624968
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009624976
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009624984
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009624992
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009625000
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009625008
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009625016
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009625024
Jul  2 03:07:58 Cyberdyne kernel: nvme_log_error: 21 callbacks suppressed
Jul  2 03:07:58 Cyberdyne kernel: nvme1n1: I/O Cmd(0x2) @ LBA 1009638744, 264 blocks, I/O Error (sct 0x2 / sc 0x81) MORE DNR 
Jul  2 03:07:58 Cyberdyne kernel: blk_print_req_error: 21 callbacks suppressed
Jul  2 03:07:58 Cyberdyne kernel: critical medium error, dev nvme1n1, sector 1009638744 op 0x0:(READ) flags 0x0 phys_seg 33 prio class 0
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636696
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636704
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636712
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636720
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636728
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636736
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636744
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636752
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636760
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636768
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636776
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636784
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636792
Jul  2 03:07:58 Cyberdyne kernel: md: disk1 read error, sector=1009636800

 

Thanks all

test.txt

[trurl]

Attach Diagnostics to your NEXT post in this thread 

[Harold_Boom]

Ah yep sorry, attached

cyberdyne-diagnostics-20230702-0951.zip

[JorgeB]

Diags are after rebooting, but it does look like that NVMe device is on its way out, these values should be zero for a healthy device:

 

Media and Data Integrity Errors:    1,034
Error Information Log Entries:      1,034

 

[Harold_Boom]

So is it worth replacing or just move everything off it and let it die? I’m just worried back my parity checks always show errors

[JorgeB]

I would replace it.

[Harold_Boom]

Ok thanks mate! I think it’s actually under warranty still so I will contact the vendor