ReadyNAS NV+V2 Problems With Disk Failure + Disk Unrecoverable Errors

[wildwood]

The unit is a ReadyNAS NV+V2 (RND4000) fitted with 4 x 3TB Seagate Barracuda ST3000DM001 Disks in a Flex-RAID (Raid 5) volume, it runs RAIDiator 5.3.11 and is powered through an APC UPS. Usage is around 60% and the unit is always shut down (wnen necessary) through RAIDiator.

I was copying files to the unit when it became unresponsive and I noted that the Disk Activity light was flashing continiously (the four Disk LEDs were lit and steady).

I left the unit for an hour and when I checked the Disk Activity Light was out, the the four Disk LEDs were lit and steady but the "Status display screen" reported that the Volume was on life support!

I checked the RAIDar display and it showed:
     Overall Status - Yellow
     RAID volume status - Flashing Red (Lifesupport)
     Disk status - Green (all four disks)

Logging in to the unit the initial screen showed an image of the unit with a green light against each disk.

The System - Overview screen showed a red warning triangle against Volumes and the notation RAID 5 (Life-Support).

The System - Info screen showed a number of oddities described below.

The "Info" section showed the four disks, each had a green light next to them but there were differences in the information provided about each disk:
  Disk 1, 3 and 4 showed the disk model (ST3000DM....) then temp, write cache and status (OK)
  Disk 2 did not show anything but the temperature, write cache and status (OK)

Looking in the "Logs" section a number of errors were reported

16:01:29 Red Disk Removal Detected [Disk 2]
16:01:29 Red A disk was removed from the ReadyNAS. One or more RAID volumes are currently unprotected ......

      Needless to say I had not physically touched the unit much less pulled a disk

16:01:43 Yellow Disk failure detected
1g:01:43 Yellow If the failed disk is used in a RAID 1, 5, or X-RAID volume, please note the volume is now unprotected .......

       It does not say which disk this message applies to, I assume disk 2

16:03:12 Yellow Detected increasing uncorrectable errors[40] on disk 4 ......

This message is repeated with the number in [] increasing each time

I would be grateful for any advice and suggestions as to the best course of action to take as it appears I have one dead disk [disk 2] and one failing disk [disk 4]. An odity is that all four Disk LEDs are lit and steady and RAIDiator shows Green against all the disks!

I assume that I need to do the following:
1. resolve the disk 2 issue first by replacing the disk and allowing a resync.
   then
2. resolve the disk 4 issue first by replacing the disk and allowing a resync.

My concern in that disk 4 with increasing uncorrectable errors may fail before the first resync so I am considering taking a new full backup of the unit before doing anything. This raises new issues such as will the data be network accessible (my B/U unit is attached to a computer) and will disk 4 last long enough to do this backup.

Thanks in advance for any help.

[wildwood · Accepted solution]

Thanks for the response.

I've managed to get the bits of data that weren't already backed up off the array and as per your suggestion am going to build a new array using WD Red drives.

Thanks again for your feedback.

[StephenB]

---

@wildwood wrote:

My concern in that disk 4 with increasing uncorrectable errors may fail before the first resync so I am considering taking a new full backup of the unit before doing anything.

I agree this needs to be a priority.

---

@wildwood wrote:

I assume that I need to do the following:
1. resolve the disk 2 issue first by replacing the disk and allowing a resync.
   then
2. resolve the disk 4 issue first by replacing the disk and allowing a resync.

---

A slight variation is to power down the system after the backup and then clone disk 4.  Use the clone instead of the original, and perhaps also remove disk 2 before you boot up.   Then insert the replacement for disk 2.  The clone needs to be a sector-by-sector copy.

---

@wildwood wrote:

The unit is a ReadyNAS NV+V2 (RND4000) fitted with 4 x 3TB Seagate Barracuda ST3000DM001 Disks in a Flex-RAID (Raid 5) volume, 

Unfortunately this particular disk model has a rather high failure rate in raid arrays (not just ReadyNAS).  I suggest that you replace these disks with a different model - going with NAS-purposed drives instead of desktop.

WDC WD30EFRX are good choices, as are Seagate IronWolf ST3000VN007.  You could also use the Red Pro and Ironwolf Pro models if you want 7200 rpm. Personally I find the Red drives to have good performance, and like the cooler temps that the slower rotation speed enables.

[wildwood · Accepted solution]

Thanks for the response.

I've managed to get the bits of data that weren't already backed up off the array and as per your suggestion am going to build a new array using WD Red drives.

Thanks again for your feedback.