Please help - Multiple Disk failures.

[Joe]

I have a red X in my array and another disk with errors.  This happened to Disk 1 (SDH) not long ago.  I did the a removed it and re-added it to the array and it rebuild fine.  I then ran a long SMART and passed with no problem.  Now a month later same problem on Disk1 with 3 errors but also Disk 2 has 23 errors.  Should I shut down the array, reboot, shut the box off?  Any assistance would be great.  I am attaching Diags and Smart reports.

Thanks

Joe

tower2-diagnostics-20210218-1337.zip tower2-smart-20210218-1345 (1).zip tower2-smart-20210218-1345.zip

[JorgeB]

Disks 1 and 2 dropped offline at the same time (and reconnected with different identifiers), because of that SMART reports are empty, but unlikely to be a disk problem with 2 disks at the same time, most likely power/connection issue, start by checking if the disks share a power cable or splitter, then post new diags after a reboot just to check SMART.

[Joe]

I just rebooted the system.  I set the array not to autostart.  Disk 1 is still marked with a red X and Disk 2 has a green dot.   I have uploaded a new SMART report for the drives.  The 2 disks do share a power connection.  I will leave the array down until I hear back.  

Thanks so much

Joe

tower2-smart-20210218-1421 (1).zip tower2-smart-20210218-1421.zip

[Joe]

Should I delete disk 1 and then put it back in and see if I can get a rebuild going?

Thanks

Joe

[JorgeB]

As expected disks look fine, if the emulated disk is mounting and data looks correct you can rebuild on top, but also good idea to try and fix the problem or it will likely happen again, try at least to replace the cables for both disks.

[Joe]

Thanks - I did have a weird issue when rebuilding this system last year.  I have 4 14TB drives 2 Toshiba, 2 WD.  The WDs would not power up of the sata power from the power supply only from a molex converted to sata.  The Toshibas run fine.  They have been running fine for most of a year that way.  Any ideas?  New power supply or go all Toshibas?

The Raid is now rebuilding but I don't want this to happen again.

[JorgeB]

17 minutes ago, Joe said:

The WDs would not power up of the sata power from the power supply only from a molex converted to sata.

That's likely the WD 3.3v SATA issue, not a reason for the failures, bad/failing molex adapter might be.

[Joe]

I don't believe these were shucked drives.  They are Western Digital 14TB Ultrastar DC HC530 SATA HDD - 7200 RPM

[JorgeB]

Just now, Joe said:

I don't believe these were shucked drives.

That issue is not exclusive to shucked drives, it's the SATA PWDIS feature, which can be present in some newer disks.

[Joe]

I never knew that!  I thought it was shucked only.  After my rebuild I will pull the drives and cover pin 3 and connect the drives to my sata rail.  Thank you so much for this info.

Joe

[Joe]

One other thing is I may impose.  Do you know why I get Current Pending Sector is 1 then a little while later Current Pending Sector returned to Normal on my cache drive.

Unraid Cache 2 SMART health [197]: 19-02-2021 08:07

Warning [TOWER2] - current pending sector is 1
CT1000MX500SSD1_2018E2A00323 (sdc)

Unraid Cache 2 SMART message [197]: 19-02-2021 08:38

Notice [TOWER2] - current pending sector returned to normal value
CT1000MX500SSD1_2018E2A00323 (sdc)

[itimpi]

It is a known issue with the firmware on Crucial SSDs that hey have stated they have no intention of fixing.

[JorgeB]

Yeah, just remove that attribute form the ones that are monitored for that device.

[Joe]

Thanks - I just removed that attribute.

[PeteAron]

Joe, it may also be worth checking out your power supply.  If it is old or not quite powerful enough, it might be a problem for you.

[Joe]

Thanks - Power supply is fine.  I just taped pin 3 on the drives as JorgeB recommended and now they work fine off the SATA rail.  Hopefully it will prevent them going offline again.