Help needed, I think my drive is dying/dead

[Moussa]

My drive mounted at /dev/sdc is showing errors and the system log is spammed up with 'program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO' errors. The SMART report seems to be gibberish so I'm not sure if the drive is dead or it's something else. Thanks for the help.

 

Running 6.0-beta12

 

Edit: After rebooting the drive is showing up as redballed but the SMART report generated now looks fine? Also ran a SMART short self-test and it passed fine.

smart.txt

syslog.zip

syslog.zip

smart_after_reboot.txt

[dgaschk]

See here: http://lime-technology.com/wiki/index.php/Troubleshooting#What_do_I_do_if_I_get_a_red_ball_next_to_a_hard_disk.3F

[Moussa]

I've followed all that information in the guide, and checked all the connections. I have the drive rebuilding now but I'm not sure where to proceed from here. Is this something to worry about? How can I figure out if the disk controller is failing, or the port is, or any of the other components in my system.

[RobJ]

I think the drive is fine, as it completely lost contact with the system, but proved fine in your last SMART report.  At Dec 24 10:31:13, the drive appears to very suddenly drop off the system, so after the kernel tried and failed to recover it, marked it as 'disabled'.  You can completely ignore all of the subsequent error messages.

 

I'd look for a loose SATA or power cable (or power splitter) to the drive, that may have slipped off, or a back frame connection that vibrated loose.  I don't believe the drive itself is at fault.  If you cannot find anything wrong at all, and it happens again, then try connecting Disk 2 to a different SATA port.

[Moussa]

Thanks for the help. I'll give everything a look over again. I don't have any spare sata ports so I may just swap over with another drive to see if the issue is actually the port.

[Moussa]

So this happened again twice. The first repeat was exactly the same situation with the same drive, but SMART looked fine and I didn't have time to do a proper investigation so I just let it rebuild.

 

Then 4 days later I see the same error in the logs but this time it's the parity drive. Obviously started getting write errors to it after the messages started appearing in the log. Drive SMART report, again, is fine after server restart. It's now rebuilding the parity drive. Please find the syslog attached.

 

So now I'm confused as to what the issue really is. It doesn't appear to be the drives themselves so what could it be?

syslog.txt

[Squid]

So this happened again twice. The first repeat was exactly the same situation with the same drive, but SMART looked fine and I didn't have time to do a proper investigation so I just let it rebuild.

 

Then 4 days later I see the same error in the logs but this time it's the parity drive. Obviously started getting write errors to it after the messages started appearing in the log. Drive SMART report, again, is fine after server restart. It's now rebuilding the parity drive. Please find the syslog attached.

 

So now I'm confused as to what the issue really is. It doesn't appear to be the drives themselves so what could it be?

It could still be cabling / power connectors.  But another suspect would be a power supply either too weak or on its way out.

 

Thanks for the help. I'll give everything a look over again. I don't have any spare sata ports so I may just swap over with another drive to see if the issue is actually the port.

Did you happen to have swapped the original drive's port with the parity drives port?

[Moussa]

Hmm the power supply might be a good shout. I don't think it's too weak (it's 430W for 6 disks and it handled them fine for almost a year) so I will have to try and RMA it.

 

No, the original drive and parity drive are on different cables and ports so I don't think that's the issue.