Drobo V2 often rebuilds/resyncs for a few minutes

Hi all,

Not sure if this is a problem (it’s annoying to me) or if its ‘normal’ Drobo behaviour.

If I’m downloading to my Drobo, while streaming a video, the lights often start to flash green and orange like its rebuilding. The video stutters so badly I have to stop it and after 5-10 minutes the lights all go green and everything is back to normal.


4 x WD 1.5TB Green
Drobo V2
FW800 using the uBCore drivers.
Windows 7 Professional 32-bit.

Any help/advice would be greatly appreciated.

it sounds like a loose connection, or a stalling drive (it stops responding for a while)

the reason the “rebuild” is so quick is that drobo has a “fast” rebuild where it can detect that its a just removed drive being re-added to the array so it just checks it is ok, rather than wiping it and rebuilding from scratch.

you dont mention what firmware you have?

regardless i’d open a support case and send in your logs an they will let you know which drive is causing the problem and why.

are you running dashboard, do you have e-mail alerts enabled? does it tell you it senses a drive has been removed and then readded?

Thanks for the quick reply Docchris.

I stupidly didn’t set the email alerts when I formatted my PC recently. I’ve just set them up now.

When I got home today I connected the Drobo via USB to see if it exhibits the same problem. So far, so good but its very random. I’m running FW 1.3.5 and Dashboard 1.5.1.

I’ll report back once its happened again.

Windows 7 and Firewire devices do not play well with each other. In fact, we have stop supporting FireWire connections on Win 7.

You probably will not see any issues on the USB but since there are issues with connectivity on FW and Win 7 I would recommend a chkdsk as soon as possible. Lost connections cause data corruption.

Thanks Jennifer. Must be quite serious if you’ve decided not to support it.

However, the Drobo just did exactly what it’s been doing even though its connected via USB. I received an email from the Drobo stating “Data protection in progress”. It doesn’t say one of the drives were removed.

What should I do? Contact Drobo support and send them the logs? Seems like the logical thing to do!

Yes, please.


I just ran my second chkdsk in 2 weeks and it never seems to complete. It gets to the end but doesn’t go back to the command prompt.

Last line reads “1236407933 allocation units available on disk.” with a blinking cursor underneath.

Are doing chkdsk h: /x (where h is your drive letter)?

I did chkdsk x: /f but it did ask me if I wanted to dismount the volume and I said yes.

What size is your volume?

I think someone posted that they had issues with Win 7, chkdsk and 16 TB volume sizes.

My volume is 8TB. I just re-ran chkdsk with the /x switch and it got to the same point.

The tech support guy says the drive in slot 3 has bad blocks which could be causing the random relayouts. He says its normal to have bad blocks on consumer drives. CHKDSK reports 0kb bad sectors so I assume that if I send this drive back to WD they’re going to say that there’s nothing wrong with it and I’ll be out of pocket?

I have to yet to hear of WD having problems with RMA’ing their drives.

Go thru website for the drive replacement:
Western Digital: http://support.wdc.com/warranty/index.asp?lang=en

Thanks for all your help Jennifer :slight_smile:

I had to RMA a WD Green 1TB drive when I first got it, as Drobo v2 marked it bad. It was right when I first got my Drobo, just about a year ago.

I was very skeptical-I didn’t think that a brand new drive, straight out of the box, would go bad like that. I thought I was in for an out-of-pocket return experience. WD RMA’d it no questions asked at no charge, and they even paid shipping (!) IIRC.

I’ve also had very good luck returning WD external HDs. I’ve had two go bad within warranty. Both times, they have given me a new, larger drive.

In my experience WD definitely stands behind their products.

My Win7 x64 machine still runs Chkdsk but don’t return to the prompt. I have a 16TB volume.
Errors are repaired by Chkdsk - it just doesn’t return to the command prompt for whatever reason.

It’s safe to just close the Command Prompt.

Unless there’s something causing more errors, a subsequent run of Chkdsk should say “Chkdsk has scanned the volume and found no errors in the filesystem.” (not exact wording)

I have RMA-ed drives to both WD and Maxtor (now Seagate) - both have been problem-free for me.
I usually get Advance Exchange (credit card deposit) whenever possible, which gives you the replacement drive and (usually) a prepaid return label. I use the same packaging the replacement drive comes in to return the old drive to ensure it’s “approved” shipment packaging.

Thanks bhiga. Pending the result of the extended test I’m busy running on the ‘faulty’ 1.5TB, I’ll probably end up using the Advance Exchange method.

However, I did buy 4 x 2TB WD Green drives today as I was starting to run low on space. This gets expensive quickly!

Advance Exchange is what I was thinking of. Thanks for reminding me! It definitely sped up the RMA process and got me a prepaid shipping label.

I think I’ve the same problem and I’m using an iMac and it happens with either USB 2.0 or FW800. The Drobo is connected directly to the iMac also. This isn’t a Windows issue and can some one explain how a loose USB/FW cable can cause a device with its own firmware on the other end to think an internal drive is bad? Seriously?[hr]

Also, why do I need to goto support to find out which drive is causing these issues? Is it really too difficult to email me that drive#X is causing a problem?

You need to send your logs to support so they can look at the “warning signs” from a drive. Not all drive events will cause Drobo to mark a drive bad (luckily, because otherwise we’d have bad drives all the time and never see green!) - minor events like timeouts can be logged, but the Dashboard doesn’t notify until the drive has reached whatever threshold is necessary for Drobo to consider it bad.

I guess the problem here is I actually had two drives doing this. The Drobo keeping quiet on a single drive makes sense, it’s one fault tolerant but when two drives are throwing errors then it should tell me straight away as two faults at the same time could mean it’s going to lose data and I’d never know till it’s too late.