Swapping (4) 1.5 TB drives to (4) 2 TB drives

can you hear the drives still working? or do they seem to have powered down

They’re still running. The progress bar and the time remaining seem to be synced right now. I’ve left dashboard running and the progress bar is @ 90% with 29 minutes left.

I have no idea if this is right or not, but it seems as if it builds some type of activity “to-do” list then has to go through that list. Any file changes are added to the end of the list. For example, I swapped a drive with 3.5 TB to sync. I then moved 1 TB off the drive. Then moved 400 GB off the drive. So even though the drobo has 50% free space, it first has to sync all the drives to the 3.5 TB state, then the 2.5 TB, then finally the 2 TB state. I know it doesn’t sync to each of those states, but I really can’t think of a better way to explain it.

Down to @ 95% done with 18 minutes to go.[hr]
Synced. :slight_smile:

Now to drive two.

If you haven’t already started the next drive, it might be worth taking a break and having support look at your logs, just in case there’s a drive in there that is “on the way out” but hasn’t exceeded the “mark me bad” threshold yet.

Relayout is a drive-stressful process and the last thing you want to happen is to have a drive (completely) fail during relayout.

agree with bhiga (and myself!)

I opened a support case and thankfully, the report was all the drives looked ok.


You ain’t kidding. A big reason why I upgraded from the 1.5 TB drives to the 2 TB drives was that I already sold the 4 1.5 TB drives. Would suck to find out that the one or more were bad.

Glad to hear your drives are OK!

Thanks. While this whole process is a pain, it is nice to see only 3 blue dots lit. :slight_smile: For a while anyway.

wow - you dont have much data on your drobo then?

I’ve pulled 2 TB off it while it was resizing. In a couple hours, I will have pulled another 500 GB of it as well. The last two drives will be syncing with about 800 GB of data. Once it has stabilized with all four new drives, I’ll copy the data back.

The second replacement drive was still at 110 hours last night. During the entire week of rebuilding, remember not even two complete drives, the progress bar would do something interesting. It looked like it would add a line of pixels then drop back. It would happen literally all of the time. I did a little searching and found this


I followed this procedure and the rebuild time of the drobo immediately dropped to 37 hours. Here’s hoping this was the solution.

Second drive finally done. On to the third drive.

well done so far :slight_smile:

#4 :)[hr]
Whew. 15 minutes for the last one. Done.

took 13 days?

It just took me about six days to swap two 1TB drives to two 2TB drives, so 13 days for all four looks consistent…

Hi, i’m considering getting a drobo.

Just wondering… if during those 13 days or so, a powercut would happen to occur, would my data be safe?

Drobo’s have internal batteries which are supposed to make them shut down gracefully in the event of a power failure. That does not mean that your O/S won’t leave your disks in a non-bootable state or lose data, corrupt folders, etc.

You should absolutely be running the Drobo on an UPS.

Your biggest concern during a long relayout is the loss of another drive while the other 3 are banging away 24/7 rebuilding the array… for 13 days straight in this case.

Your data is never “safe” if you only have one copy, no matter where you store the data. And that has nothing to do with the reliability of the Drobo per se. It is, like all storage, less than “fail safe”.

(You might want to track down the origin of the expression “Fail Safe”. As I recall, it originated (at least in popular culture) with a book - then a movie - of the same name written in the early '60s. It was a (supposedly fictional) system of managing the US nuclear deterrent capability that absolutely, positively could not ever fail. Until it did, with consequences that are worth reading the book or watching the move.)

The risks during a powercut are that the filesystem may be corrupted but unfinished writes (thats an oversimplification but its close enough) - this is the same on all disks - internal, external, drobo…

whether drobo is doing a relayout or not, that risk will stay the same

for drobo’s own internal filesystem - which it uses when it is actually doing the relayout - that is protected by a small backup battery