[1481082.784899] ata6: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
[1481087.792922] ata6.00: qc timeout (cmd 0xec)
[1481087.792934] ata6.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[1481087.792965] ata6.00: revalidation failed (errno=-5)
[1481088.107288] ata6: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
[1481095.385983] ata6.00: configured for UDMA/133


We can clearly see from the errors above that something is going wrong with ata6 or perhaps port "SAS/SATA port 6" but which might that be?

Use this handy script to find out

ATAPORT=ata6

ls /sys/class/ata_port/$ATAPORT/../../host*/target*/*/block/

If you don't have a nice enclosure that makes it clear which bay has which drive, you'll need to physically check inside the server or chassis.

In my case it said "sdj" so the next step is something like this:

smartctl -a /dev/sdj|grep -i serial



This way we'll know we've checked/replaced the correct drive by matching the serial.

In my cases like this actual example though the problem could be a power problem, or even more likely a bad SATA/SAS cable.

Here is the full quick script to find the drive that causes the error and get its serial:

Remebmer to change ata6 to the ata# of your error.

ATAPORT=ata6
drive=`ls /sys/class/ata_port/$ATAPORT/../../host*/target*/*/block/`  
smartctl -a /dev/$drive|grep -i serial

dev, ata, sata, gbps, sstatus, scontrol, qc, timeout, cmd, xec, err_mask, revalidation, errno, configured, udma, errors, quot, sas, ataport, ls, sys, ata_port, enclosure, ll, server, chassis, sdj, smartctl, grep, ve, replaced, matching, remebmer,

Latest Articles