How to find the cause of "Input/output error" errors on Linux?

Question

linux

My utility (verifies file hashes) works fine on various Linux servers, but on one of them it regularly crashes with the "Input/output error" error.

Conclusion dmesg --level=err:

[    1.329324] ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PCI0.SAT0.SPT0._GTF.DSSP], AE_NOT_FOUND (20200925/psargs-330)
[    1.329375] ACPI Error: Aborting method \_SB.PCI0.SAT0.SPT0._GTF due to previous error (AE_NOT_FOUND) (20200925/psparse-529)
[    1.336931] ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PCI0.SAT0.SPT0._GTF.DSSP], AE_NOT_FOUND (20200925/psargs-330)
[    1.336980] ACPI Error: Aborting method \_SB.PCI0.SAT0.SPT0._GTF due to previous error (AE_NOT_FOUND) (20200925/psparse-529)
[ 3013.266044] ata1.00: exception Emask 0x0 SAct 0x18 SErr 0x0 action 0x0
[ 3013.266054] ata1.00: irq_stat 0x40000008
[ 3013.266059] ata1.00: failed command: READ FPDMA QUEUED
[ 3013.266066] ata1.00: cmd 60/00:18:98:db:b8/01:00:44:00:00/40 tag 3 ncq dma 131072 in
                        res 41/40:00:60:dc:b8/00:00:44:00:00/40 Emask 0x409 (media error) <F>
[ 3013.266072] ata1.00: status: { DRDY ERR }
[ 3013.266075] ata1.00: error: { UNC }
[ 3013.281961] blk_update_request: I/O error, dev sda, sector 1152965728 op 0x0:(READ) flags 0x80700 phys_seg 2 prio class 0
[ 3956.517696] ata1.00: exception Emask 0x0 SAct 0x18 SErr 0x0 action 0x0
[ 3956.517706] ata1.00: irq_stat 0x40000008
[ 3956.517710] ata1.00: failed command: READ FPDMA QUEUED
[ 3956.517717] ata1.00: cmd 60/00:18:90:99:f6/01:00:4e:00:00/40 tag 3 ncq dma 131072 in
                        res 41/40:00:78:9a:f6/00:00:4e:00:00/40 Emask 0x409 (media error) <F>
[ 3956.517723] ata1.00: status: { DRDY ERR }
[ 3956.517726] ata1.00: error: { UNC }
[ 3956.533817] blk_update_request: I/O error, dev sda, sector 1324784248 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 0
[ 3961.281636] ata1.00: exception Emask 0x0 SAct 0x1 SErr 0x0 action 0x0
[ 3961.281645] ata1.00: irq_stat 0x40000008
[ 3961.281649] ata1.00: failed command: READ FPDMA QUEUED
[ 3961.281657] ata1.00: cmd 60/08:00:78:9a:f6/00:00:4e:00:00/40 tag 0 ncq dma 4096 in
                        res 41/40:00:78:9a:f6/00:00:4e:00:00/40 Emask 0x409 (media error) <F>
[ 3961.281664] ata1.00: status: { DRDY ERR }
[ 3961.281667] ata1.00: error: { UNC }
[ 3961.297276] blk_update_request: I/O error, dev sda, sector 1324784248 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0

What could be the reason?

Answer 1

The reason is that a wonderful furry animal scribe digs a mink on the disk.
134 sector move events, 152 moved sectors and 1222 sectors pending verification. You can ask Zettabyte - is this a complete scribe or just an ordinary one? :)

Answer 2

The hard drive is covered. See SMART
Most likely, a replacement.

Answer 3

carrier problem.
more often - a cable or overheating of the south bridge.
less often - they wrote above the bad sector.