Drobo

Drobo working drives to death all times of the day.

Hey all,

I just got a v2 firewire drobo, threw in 4 2tb hitachi deskstar drives. But the noise is ridiculous! After I plugin my drobo on my 27″ iMac, it sounds like it’s writing tons of data between the drives, working them to death all day long. And this happens every time my computer wakes up from sleeping, or if I turn on and plugin my drobo.

According to the activity monitor, the computer is not writing anything to the drives. Time Machine is not using the drobo. I thought it might be spotlight cataloging everything, but this happens EVERY TIME my machine wakes up or restarts.

According to drobo tech support everything is peachy. No problem whatsoever according to the Drobo Diagnostic log I sent them.

Anyone have any problems like this? Any suggestions/recommendations?

THANKS![hr]
Oh 1 more thing. My drobo is about 70% empty. So it’s not the ‘5-10% space remaining’ copying issue I’ve seen in other posts.

If you disable spotlight does it still happen???

I haven’t hard disabled spotlight via the terminal. But I have left my computer and drobo on for days, without writing any new data to the drobo. So I don’t see what there is to catalog, nonstop, 24/7. And oddly according to the disk activity monitor, when my computer is just sitting idle, and my drobo is cranking away on the drives, there is no I/O disk activity being shown.

I don’t really want to have to disable spotlight either. Unless it causes some known problem with the drobos.

cool Chameleon BTW Docchris.

I just did a test, completely disabling spotlight, and still my drobo is working the drives to death, like they are copying tons of data. So now I’m completely stumped.

To be 100% certain from a terminal run
sudo lsof | grep “/Volumes/Drobo”

This will list any open files on the Drobo. If spotlight or some other app is reading or writing a file it will show up.

I get

fseventsd 55 root 8u REG 14,5 0 73859 /Volumes/Drobo/.fseventsd/63657336c3d024e5

I’ve told spotlight not to access the DROBO, but it’s still churning away all day and all night !

Probably the simplest test is to disconnect the Drobo from the computer. If it’s still running the drives, then Drobo is doing something. If it isn’t, then the computer’s having it do something.

How long should it be ‘doing something’ for ?

That all depends. If you’ve recently added a drive, it’ll rearrange in the background. I don’t know how long the background relayout takes. In theory it should be the same or less time compared to a drive replacement relayout, which is dependent on how much data has been added or removed.

It’s a 4Tb array, I added 1.8Tb to the existing 800Gb, then deleted the original 800Gb.

Well, when I went 8 TB -> 6 TB -> 8 TB, it took somewhere in the upper 60 hours to complete the relayout.

I think a feature needs to added to the control panel to indicate when this activity is taking place and how long it is going to take.
It does seem to affect transfer times and the constant activity must be having an effect on the lifetime of the drives.

Drobo Dashboard tells you when relayout is happening and gives an estimate on completion time.
Well, actually it tells you when it’s a “critical” condition. If it’s a strict drive addition, a “silent” relayout is done and I don’t think it tells you that it’s doing anything - or how long it’ll take.

EDIT: Clarification…

The Drobo is probably doing some internal housekeeping. I have measured a great loss of performance hours after formating a volume. That housekeeping does stop if it is not connected to an active computer. Keep the computer on for a day or two and see what happens.

From my point of view, it’s preventing me from using it for it’s intended purpose.
I use the Drobo to store HD content which is streamed over a network.
While this ‘housekeeping’ is taking place, it’s causing 1080p content to stutter as the transfer rate is varying so wildly.

It’s now been housekeeping for 10 hours a day for nearly a week.

How long do I wait before I can use my Drobo ?

If this is the outcome every time I transfer a large amount of data I might be forced to find another solution.
I just can’t afford this amount of downtime.
At least if I had an indication that there is some actual progress or a percentage complete indication, I could find a temporary solution in the meantime, but I’m completely in the dark.
When you need the device to fulfill a function, being told to just ‘leave it’ does not offer much comfort.
If it only took a few hours, then that would be acceptable, but I’ve essentially lost use of the Drobo for a week because I transferred some data !!!

Madness.

Once the housekeeping is done, it “just keeps up” until you add more storage and it needs to do more housekeeping. If you don’t add more storage, it goes as it needs to.

What kind of bitrate are you streaming? 1080p is just the resolution, the bitrate is really the key.