0

we have linux redhat server version 7

we are suspects that one of the disks (/dev/sdc ) in the machine HW ( dell ) is in bad state, ( and that because required to do fsck from dmesg )

but we not sure , machine is on new HW and new disks

we found the utility - smartctl

and from man page seems that (-H) can summary the health status

smartctl -H /dev/sdc

in case we get from the smartctl the following:

SMART Health Status: OK

is it enough to get the disk status so we can be sure that disk is ok?

shalom
  • 451
  • 12
  • 26
  • Please show the output of `smartct --all /dev/sdc` and add detailed hardware information (ie: server type, RAID card in use, ecc). – shodanshok Nov 12 '17 at 15:52

1 Answers1

2

As I say in more length elsewhere, the best data we have on the subject tell us that smartctl is a good predictor of failure, but not a good predictor of health. In other words,

  • if SMART tells you your drive is failing, then it almost certainly is, and you're an idiot if you don't get your data off it as soon as possible. But,

  • just because SMART doesn't tell you that your drive is failing, that doesn't mean it's not on the way out.

You have good reason to believe a drive is on its way out. Replace it as soon as is humanly possible, and check your backups are good. Just because smartctl doesn't yet see failure in it does not make it a good drive.

MadHatter
  • 78,442
  • 20
  • 178
  • 229
  • ok ,fine regarding that what are the other options that indicates disk is in bad state . or is it possible to identify this by CLI linux commands ? – shalom Nov 12 '17 at 15:45
  • You're missing the point. You're still looking for a definitive indicator that the disc is going bad. I say you won't always get one, so waiting for it leads to data loss. If you have grounds for thinking the HDD is going bad, then don't take chances, just replace it. – MadHatter Nov 13 '17 at 01:36