we parse the SMART output from the CLI response, so if the command line tool reports passed it will show up as passed.
what do you get if you run the same command in synology?
however the drive is problematic regardless of the passed status.
EDIT: found something:
@oguz please check this:
smartctl -H /dev/sde return exit code 0
smartctl -a /dev/sde return exit code 192
My guess is that Proxmox use the -H to check the exit code wich actually does not seems to work great and should be replaced with -a.
ORIGINAL POST:
Still digging into this and i think i may have a school case that may need to be checked and eventually corrected:
So, this drive /dev/sde shows as SMART passed but fails at extended self test at 90%
I know how the exit codes are used and how they work using 8 bits.
I use this code to check for bit values provided by the man page of smartctl at
manpage
Code:
#!/bin/sh
smartctl -a /dev/sde > /dev/null
varstatus=$?
for ((i=0; i<8; i++)); do
echo "Bit $i: $((varstatus & 2**i && 1))"
done;
Here's the output for a drive /dev/sdc wich has been a bit hot once and it does match exit code, bit value and smart explanation for this bit.
Correct me if im mistaken but the bit define the exit code so there may be thoses values and they may be combined.
0 1 2 3 4 5 6 7
1 2 4 8 16 32 64 128
same test for the failing drive reports this:
This would match the description of the smartctl exit codes bits wich are:
bit 6: Device error log contains records of error
bit 7: Device self-test log contains records of errors ......
What do you think about this? This drive should be reported as failing but is not - How?
Regards - JS
