Drobo

Lost redundancy after replacing disk

Hi all,

I have a strange situation. Because of this, I now have NO REDUNDANCY, so any help is appreciated…

In my DroboPro I had 5x 1TB, 1x 1.5TB and 1x2TB. All drive lights were green, single-drive redundancy and (only) 1.5TB of data on the (single) volume.

So I decided to remove the 2TB drive, and replace it with a spare 1TB drive. Obviously, when I removed the 2TB, drivelights started blinking green/orange, but soon after stopped. Then all occupied drive bays were solid geen, and the left-most empty bad was solid red. DroboPro’s standard way of asking for an extra drive.

Question 1: Why would it need an extra drive if I have 1.5TB data and 5x 1TB + 1x1.5TB? I had single-drive redundancy, but even on duad-drive redunancy I should have 4TB available. By far enough for my 1.5TB, no?

So I added the 1TB drive, which made it 6x 1TB + 1.5TB. But still it’s asking for another drive (in the 8th bay).

I have the feeling it has to do with the 2TB which was the biggest drive in the set, that I removed. Maybe the 0.5TB extra space that only this drive had, now needs to be replicated to other disks. But then I would expect green/yellow blinking lights.

Questions 2 & 3: What’s happening, and how can I fix this so I have redundancy again?

Thanks,
EgMaf

According to the calculator: http://www.drobo.com/calculator/8-bay/index.php your initial setup had 5.9TB of available storage, 1.37TB used for protection and 0.46TB reserved for expansion.

Remove your 2TB and that changes to 4.35TB available, 1.09TB used for protection and 0.46TB reserved for expansion.

As you say, it looks like you have more than enough storage in both cases, even if you had dual-disk redundancy turned on.

I think the answer to your question 1 (and 2) is that something inexplicable went wrong. Hopefully it’s just software and not a hardware issue. I’d copy your data off the Drobo (onto the 2TB drive you removed maybe?) ASAP! To set up again (question 3), put the disks you want into your Drobo, and set it up again (reformat, maybe even do a hardware reset). Do any firmware updates while there’s no data on there and do some testing on the Drobo while all your data is sitting safe somewhere else.

Question for you though: why did you get a DroboPro and NOT use dual disk redundancy when you have so much space and so little data?

Ow, but that question is simple (any maybe a bit stupid too…). Because of the disappointing speed of the DroboPro (on iSCSI to Windows Server 2008R2, dedicated Intel NIC), I am migrating the live data off the DroboPro onto disks internal to the server. So I started with much more data and dual redundancy, and slowly removed disks from the DroboPro, put them into the server, copy data over, wait for the DroboPro to finish its automatic redundancy build, remove another disk etc.
To make the 1st disk available, I needed to switch from dual-disk redundancy to single-disk redundancy…[hr]
Eh… the question wasn’t stupid, but as you can read in the answer, I might have been… just a bit… ?

Ok, Seems like I have the same problem. Deleted a lot of files from the DroboPro, All In all I am still using 1TB of space. However I have 2x1Tb, and 1x3TB drives in the Drobo. The red light in bay five is on. If I add another drive, say a 1Tb, it does not begin to relay out, but rather a green light appears underneath this new 1TB drive, and the red light moves over one space, to light up bay 5!

According to the Dashboard: “Drobo cannot protect your data against hard drive failure until you increase total capacity”. What on earth is going on?

Since you’re moving a lot of data off Drobo, you should be aware that Drobo’s garbage collection can take some amount of time to catch up. This will be apparent by the capacity lights and what Drobo “thinks” you have in use.

It doesn’t affect the data integrity - connected clients will see the proper amount of used space, but it can affect the internal housekeeping functions of Drobo.

Likely this is why Drobo wanted you to add another drive. It probably thought you had way more than just the 1.5TB you have left on it. Leave it be overnight and wait for the garbage collection to catch up.

I recently did a mass-delete (2 TB worth, approx 45% of my total protected storage) on my Drobo v2 and it took 5+ hours before it completely freed all the space internally.

It’s not normally a problem, but definitely can be in mass-move cases like yours, and IIRC, Docchris got into a really bad situation where there was supposed to be free space available, but Drobo hadn’t “caught up” yet and it somehow got into a chicken-and-egg loop.

So definitely for large deletes/moves - do them in small chunks, let it rest and recover in between.

I apologise, but, in my opinion, you are mistaken. I can prove it. Write to me in PM, we will discuss.