When is it safe to replace the failed drive in this instance?

Drobo - End of Sale Drobo 5N
1

I have a Drobo 5N with the latest firmware 3.5.11 [8.90.81921]

It used to have 3x3TB and 2x4TB installed. In May I replaced two of the 3TB drives with a pair of 6TB, and left a 3TB and two 4TB drives in it (so now 2x6TB, 2x4TB, 1x3TB). I first backed it up to a QNAP NAS with 4x6TB drives, so I would not lose data. At that time I installed the new firmware that would support 64TB and wiped/reset the Drobo back to default factory settings, and then restored all my data from the QNAP after re-creating all my shares.

Everything was fine until tonight, 4 months later, when I get an email alert that “Data protection is in progress.” I check the drobo dashboard and one of the new 6TB drives in Bay 0 is “missing”. I go down to the basement and look at the drobo, and bay 0 is solid yellow, and bay 1-4 are blinking green/yellow. I double check the drive in bay 0 and it seems to be tightly inserted, therefore it’s dead and the drobo just doesn’t see it anymore.

While it was working on Data protection, I wanted to remove the bad drive and insert a new one, but the second error message said “do not remove any drives”. So, I waited a couple of hours and when it was done, the drive in Bay 0 LED is now red, and the others are green, as I type this.

I have 11% free space with just the 4 drives working. I know I’m ready now to install my backup 6TB drive that I wisely bought a while back.

(1) Could I have installed the 6TB drive in Bay 0 while it was doing Data Protection In Progress? Need to know for next time.

(2) Why does the Drobo now say “Drobo cannot protect your data against hard drive failures until you provide at least 2 hard drives for single, or 3 hard drives for dual disk redundancy. Insert a new hard drive into the empty drive bay indicated by the red light.”

Shouldn’t I only need to insert one, not two drives?

(3) As I was typing this post, it keeps switching between the message in #2 above, and then trying data protection for a 30-60 seconds, and then back to message #2, and I get emailed EVERY TIME! Why does it do this if I haven’t inserted the drive yet. I want to insert a 6TB disk now, but not if it switches to data protection in progress while I’m doing it, if that could cause data lose.

What do I do?

2

(1) I see no danger in removing the failed drive during data-protection…it is already “removed” as far as Drobo is concerned.

(2) Sounds like you have single-disk redundancy set instead of dual-disk redundancy (preferred), so you don’t have ANY redundancy right now.

(3) Inserting a drive now is the only way you will get any type of protection. You should make sure your backups are current and replaced the failed drive. If another drive fails during the grueling rebuild process you may lose your files.

3

Thanks. After the drive failed I eventually got the solid red light on Bay 0, indicating that I could replace the drive. The FAQ said that during the relaying process with flashing green/yellow lights, that I should not remove any drive. It didn’t say I could remove the solid yellow drive. When I was starting the new thread, the relaying process (rebuild) was finishing just before I finished my post, with 1 red LED and 4 solid green LED for the 4 drive bays showing up after about 2 hours.

Once the Bay 0 drive LED was solid Red and the other 4 showed solid green LED (after it did a 30 second check for data protection again), I quickly popped out the bad drive and popped in the new 6TB drive, and it started to rebuild again. All 5 drives were blinking green/yellow while it was relaying/rebuilding the array as I went to bed. This AM, when I woke up 6 hours later, I was back up to 15.43TB of storage with 13 hours remaining to achieve data protection.

– BIG CONCERN REMAINS HERE –

I’m confused because of the unusual error messages it gave to me on how many drives to add for data protection, with the wrong number of TB of available capacity. Basically - After the 1st rebuild with 4 green LED drives remaining (and one red LED on Bay 0), it was asking me to insert 2 drives for single redundancy, or 3 drives for dual redundancy! See attached photo.

The bigger problem is that with 4 solid green LED after the first data protection was complete, it was telling me I had no data protection and only 9.98TB of available capacity, not the original 15.43TB with no protection that it should have indicated (sorry, no screenshot). The 9.98TB available capacity would only be correct if the 4 drives were working with single disk redundancy. In that case it wouldn’t have asked for any new drives for data protection. Although I didn’t have dual disk redundancy turned on, I had plenty of empty space to do the rebuild with only 4 remaining drives and still have data protection on, yet it was asking for 2-3 more drives for protection when I had only 1 available drive bay! It was very confusing. Why didn’t it still show 15.43TB available capacity with no data protection, and just ask me for 1 more drive?

According to the Drobo capacity calculator: 2x6TB + 2x4TB + 1x3TB = 23TB total capacity = 20.91 Raw Capacity = 15.43TB Available Capacity (with 5.48TB data protection). I was only using 8.78TB. If a 6TB drive fails using 1 disk redundancy, then logically the raw capacity should either (1) stay at 15.43TB with no protection if the array was fairly full, or (2) be rebuilt as a 9.98TB 4 drive array with 1 disk redundancy, since I had enough free space to do that (11% free after the rebuild).

I didn’t get either one of the above - I got the smaller available capacity + no protection. What would have happened if I was using most of the 15.43TB of available protected storage when the drive failed? Would it have remained at 15.43TB with no protection, or would I have lost data since it says it dropped to an incorrect 9.98TB with no protection?

To clarify - before the first 4-drive relaying/rebuild was completed, during the “data protection in progress” message, it showed I had 9.98TB available and 1TB was free (11%) with 4 working drives + NO data protection. It never showed that I still had the original 15.43TB available with 4 drives + no data protection as it should have.

Something isn’t right.

  • I ordered another 6TB WD Red as a spare, to have like before in case this happens again. I’ll get the old 6TB drive replaced under warranty, but in reading reviews people complain that the WD sends a refurbished drive that fails fairly soon afterwards. Not sure what I’ll do wth it.

** The firmware would be fantastic if you could have single disk redundancy turned on, so that you could potentially use as much storage space as possible, yet have the array rebuild to single disk redundancy with the remaining 4 disks after a disk loss IF you are not using more space than that would allow.

4

Nothing?

5

This is a known issue. It is a compatibility issue with 6TB drives, at least with the 6TB WD Red drives, maybe others too.

It is very likely your 6TB drive ISN’T bad. Another issue is that sometimes it detects them as SSD drives and requires more “SSD” drives for proper redundancy.

My B810n had this happen, and would alternate between detecting and not detecting 6TB WB Red drives. Once it booted up and said 3 of my 6TB drives were missing. I thought all of ny data was gone, since even the unit’s name didn’t appear on the Dashboard. I rebooted as many times as needed (some had to be manual since the unit didn’t respond) until at least 2 of the drives were recognized. I inserted a 4TB in the. Empty" bay that had the 6TB drive in there, let it rebuild, then bought a Synology nas and more drives and copied everything over. The same “bad” drives work in the Synology. Some had to be manual since the unit didn’t respond) until at least 2 of the drives were recognized. I inserted a 4TB in the. Empty" bay that had the 6TB drive in there, let it rebuild, then bought a Synology nas and more drives and copied everything over. The same “bad” drives work in the Synology.