Are my drives okay?

Nan123 · 2022-07-27 23:58:14

Hi, Sorry if this is a dumb question, but I wanna make sure my drives are okay.

Here is the smart data for both of them.

I got a few errors in dmesg:

I/O error, dev sdc, sector 4223810488 op 0x0:(READ) flags 0x80700 phys_seg 17 prio class 0

I/O error, dev sdc, sector 4223811000 op 0x0:(READ) flags 0x80700 phys_seg 17 prio class 0

I/O error, dev sdc, sector 4295810776 op 0x0:(READ) flags 0x80700 phys_seg 33 prio class 0

I/O error, dev sdc, sector 4295811288 op 0x0:(READ) flags 0x80700 phys_seg 33 prio class 0

I assume these are bad sectors. There are also these errors, but I'm to dumb to decode them.

[129721.552043] ahci 0000:02:00.1: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x000f address=0x40c23db800608080 flags=0x0030]
[129758.173394] ata5.00: exception Emask 0x0 SAct 0x3006000 SErr 0x0 action 0x6 frozen
[129758.173406] ata5.00: failed command: READ FPDMA QUEUED
[129758.173409] ata5.00: cmd 60/00:68:b8:3b:c2/02:00:fb:00:00/40 tag 13 ncq dma 262144 in
                         res 40/00:01:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[129758.173418] ata5.00: status: { DRDY }
[129758.173421] ata5.00: failed command: READ FPDMA QUEUED
[129758.173423] ata5.00: cmd 60/00:70:b8:3d:c2/02:00:fb:00:00/40 tag 14 ncq dma 262144 in
                         res 40/00:01:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[129758.173431] ata5.00: status: { DRDY }
[129758.173435] ata5.00: failed command: WRITE FPDMA QUEUED
[129758.173436] ata5.00: cmd 61/08:c0:50:17:c4/00:00:d1:01:00/40 tag 24 ncq dma 4096 out
                         res 40/00:01:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[129758.173444] ata5.00: status: { DRDY }
[129758.173447] ata5.00: failed command: WRITE FPDMA QUEUED
[129758.173448] ata5.00: cmd 61/08:c8:00:82:7d/00:00:fa:00:00/40 tag 25 ncq dma 4096 out
                         res 40/00:01:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[129758.173456] ata5.00: status: { DRDY }

[129997.324962] ahci 0000:02:00.1: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x000f address=0x400ce0d800608024 flags=0x0030]
[130027.489396] ata5.00: exception Emask 0x0 SAct 0x6 SErr 0x0 action 0x6 frozen
[130027.489405] ata5.00: failed command: READ FPDMA QUEUED
[130027.489407] ata5.00: cmd 60/00:08:d8:de:0c/02:00:00:01:00/40 tag 1 ncq dma 262144 in
                         res 40/00:01:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
[130027.489415] ata5.00: status: { DRDY }
[130027.489418] ata5.00: failed command: READ FPDMA QUEUED
[130027.489420] ata5.00: cmd 60/00:10:d8:e0:0c/02:00:00:01:00/40 tag 2 ncq dma 262144 in
                         res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[130027.489427] ata5.00: status: { DRDY }

I was moving lots of data between drives (one of which is NTFS) and then dd'ing sdb this happened, data is backed up and seems fine, but just wanna be sure. I assumed that the I/O errors were bad blocks, but the smart data, if I'm reading it properly, displays no bad sectors. I also just got the sdc drive a month of so ago, so I'm a little nervous. Data is backed up.

Last edited by Nan123 (2022-07-27 23:58:54)

arch_jsb · 2022-07-28 00:05:43

I always check the Reallocated Sectors Count. those are each zero, any problem was probably a one-off heat instance, drive warmup, or flaky cable problem.

Nan123 · 2022-07-28 00:43:25

Figured as much, I'll probably switch the cable to be safe. Thanks.

seth · 2022-07-28 06:45:28

194 Temperature_Celsius     0x0002   116   116   000    Old_age   Always       -       56 (Min/Max 11/61)

is too hot for a spinning device and

194 Temperature_Celsius     0x0022   105   101   000    Old_age   Always       -       47

is very hot, but there're no recorded damages (yet)

The errors are on the bus, so either the heat or the cable (or the cable because of the heat ;-)

Do something about the temperature (it can vary drastically inside an ATX case and is most likely picked up from there - in doubt add active cooling, ie. a fan) and re-check the cable seating (pull and re-plug them)
You should also run long smart tests to make sure the heat hasn't caused any permanent damage to the drive.

arch_jsb · 2022-07-28 18:19:52

Can you suggest how long the long test should actually take?... never mind, I found the answer: it probably varies by drive and the time that it would take is given within the output of

 smartctl -c /dev/[drive]

Last edited by arch_jsb (2022-07-28 18:24:44)

seth · 2022-07-28 18:26:00

I meant "smartctl -t long /dev/sdc" - the duration depends on the size and throughput of the drive as well as possible side-load (ie. you using it), but it's typically not more than a couple of hours.

Edit: crosstalk

Last edited by seth (2022-07-28 18:26:32)

Nan123 · 2022-07-29 04:42:47

Hm, Okay. My Case is loaded up with fans, I'll see whats going on with airflow.

Arch Linux

#1 2022-07-27 23:58:14

Are my drives okay?

#2 2022-07-28 00:05:43

Re: Are my drives okay?

#3 2022-07-28 00:43:25

Re: Are my drives okay?

#4 2022-07-28 06:45:28

Re: Are my drives okay?

#5 2022-07-28 18:19:52

Re: Are my drives okay?

#6 2022-07-28 18:26:00

Re: Are my drives okay?

#7 2022-07-29 04:42:47

Re: Are my drives okay?

Board footer