Drobo

Does this rebuild time seem excessive?

Have 8x2TB WD Green drives with dual-disk redundancy on. It was probably about 70% full.

Just lost one of the drives, popped in a replacement and it wants about a week (162 hours) to rebuild. Does this seem excessive?

its not very accurate at first, it gets better

the more you use it, the slower the rebuild is too.

also i think dual disk redundancy does rebuild a bit slower than single disk redundancy (even with only a single disk failure).

it seems a little long, but not totally unreasonable, i would have said more like 48-72 hours would be what i was expecting.

I haven’t been on the forum for a long time, so I apologize in advance if I should have started a new thread instead of tacking my question onto this one.

I also have a DroboPro with 8X2TB WD Green drives with dual-disk redundancy on. It is currently about 60% full. Out of the blue yesterday afternoon, all the drive lights started flashing green and yellow/orange but then settled to solid green after a short time. I know that the DroboPro will sometimes do a relayout if there are bad blocks, so I thought that that was all it was. It did this 4 more times before I left work yesterday. It started to do it again just now, but I happened to be looking right at the DroboPro when it did. For just a few seconds, the far left drive light went red before it started flashing green and yellow/orange. I don’t have any extra drives on hand to swap out, and I hate to waste this one if it’s just the DroboPro being wonky. I couldn’t find anyone else on here having this exact problem, so I thought I’d put this out there. Would you recommend pulling this drive and seeing if that stops the repeated relayouts?

Re-allocating bad sectors is done silently in the background.

It will be that one drive/slot causing issues.

Power down and re-arrange the drives, to see whether it is that drive or slot causing problems.

Also send your log files to DRI and they will tell you why it is failing

Do not generate a diagnostic file if the drobo is in relay out, blinking green/amber. Wail until all the drives are green.

Thanks! Will do.

I thought I’d post the response from tech support, in case it helps anyone else. I sent them my diagnostic file at the end of the day yesterday, and I got a response first thing this morning. Thanks to Jennifer and Sky for the quick support!

From Sky:
"Thank you for contacting Data Robotics technical support. You’ve indicated that your DroboPro has been going into spontaneous relayouts (green/yellow flashing lights).

It’s normal for the Drobo to show a solid red light on the first drive bay when all the drives are the same size, if it’s having a problem with the drive set. The diagnostic file shows that the drive in the fifth bay in from the left has bad blocks and has stopped responding at least twice, which is what has been causing the spontaneous relayouts. The Drobo will eventually fail this drive; it would be best to replace it before that happens. The drive should still be under warranty; the warranty section of Western Digital’s website can be found at http://support.wdc.com/warranty/index_end.asp?lang=en or they can be contacted at (800) 275-4932."

good to know

Why isn’t the option greyed out then while a relayout is in process?

Because there may be occasions when it is necessary.

It won’t do any damage, but it will cause the relayout to restart

At least Sky told you which drive was failing. My rep just told me which drive serial number it was and wanted me to pull all the drives out one by one and check myself, which I found ridiculously poor customer service, but the drive had a hard failure entirely before I even had the chance to check them all manually.

I have exactly the same setting you have goodcow, and just lost a drive yesterday (about 70% used too). It has been more less 24h already, and it’s telling me that it has “11 hours remaining”. It doesn’t seemed to be very precise about it, even though I’m not using it.

Well mine did end up taking about a full week to rebuild.

Didn’t used a stopwatch, but it took something like 36h.

sounds pretty much dead on right, i usually tell people 24-48 hours…

Wouldn’t it be nice if Dashboard or an unencrypted log file told you this, instead of relying on DRI to decrypt the log files?

Indeed.

It’s a waste of monetary resources in support costs.

I agree too, as I would like to know what was wrong with the drive. I have performed three 7-Pass erases on it as a standalone drive, copied Gbs to it et al. and it’s working fine.