Nice Drobo B800Fs - happily working. No problems. Getting full. I pull a drive and put in another. Estimated time to resync - 68 hours. Ok, no problem.
After about three days of green and orange flashing lights, It is now solid orange, and though I can ping the device, dashboard won’t connect and I can’t browse to the shares any more.
Over the long term, this seems to happen over half the time a new drive is inserted - 60% yo 75% of the time.
Why must Drobo be so fragile? Why am I afraid every single time I try to add more space?
No WAY I can put this into a corporate production environment![hr]
For added info, I have experienced this same phenomenon on the following:
Drobo - original - 2 of them
Same accross the board. None of the five could be relied on to finish even half of the synchronizations that occurred when putting in a new drive, once they had a substantial amount of data on them ( greater than 1TiB)
I don’t think I have a lemon. I think there’s something fundamentally wrong. This is every Drobo I’ve ever owned, or ever helped someone else with.
Don’t know what’s wrong there, but my 4-bay Gen 2 Drobos were 85-90% full when I upgraded from 4x2TB to 4x3TB and they succeeded. Granted, I had shut down the machine that was connected to it for the duration.
I had Drobo v2 and Drobo S. Loaded with 3-4TB of data. I have upgraded multiple disks from 4x1TB -> 4x2TB in the Drobo v2, and from 5x2TB -> 4x3TB+1x2TB in the Drobo S. I may be only a data set of one, but I have not had the same experience with multiple disk upgrades with a diskpack loaded with multiple terabytes of data.
I am not attempting to deny or refute your problem, but putting forth that yours may not be the typical experience.
But, I don’t run Drobo Win2008 servers, so maybe that makes a difference.
… and around we go again. This time, I had drive #1 go bad. Red light. So I pulled the drive, and put in a replacement. 4 or 5 minutes later, the DroboFS just “went green”. No resync. No flashing yellow lights for days. Just “green”. I assume this means it “thinks” it has resynchronized, while there is actually a drive in there with nothing on it.
… and now, when copying files to the drobo, it sometimes just stops communicating. I can still ping it, but it no longer allows access to shares or dashboard. I do a clean shutdown with the power button on the back, wait a while, and start it up again. It works for a while, and then down it goes again.
… and of course, even though I have the email alerts set up and tested, I get nothing. No alert of any sort.
do you have any more info about the drives you have been using?
i ask that because i have always been sticking to WD drives of a certain Green range, and havent had a problem here (apart from power outage or windows crashes or the odd glitch every other year)… but i only use USB direrectly attached models…
if you feel they are fragile, then from what you mentioned above, youre probably justified to feel that way… but if you can, please continue to give us more info about programs and services that might be running too and maybe we can try to help pinpoint a common factor in your scenario to help improve it… (hopefully we wont find the “drobo” as that common factor) lol - but lets see
No I haven’t. But that would mean every single drobo I’ve had or been involved helping with (about 6 or 7 of them), has a “bad chassis”, because they all behave like this. I use mostly Seagate drives. I used Seagate LP 2TB drives for the longest time, cuz they run cooler. Upgrading to 3TB I’ve used Seagate 7200.12 drives.
Corndog, I would be just as frustrated in your situation. Something is obviously not right.
I don’t really have ideas for you beyond what Paul and bhiga suggested. I’ve just never seen that any time I’ve replaced a drive in my Drobo v2, Drobo S (gen2), or DroboPro. I use a mix of Seagate 7200RPM drives and WD RE and Green, and have replaced perhaps 9 drives in total. The procedure is pretty simple, so maybe our units are different somehow.
hi corndog, how popular is the data on the drobo?
eg is it just you who is accessing it, (or you and a few) or is it used by lots of people.
if its lots of “users, (and potential programs”), then as a test, maybe you can disconnect it from the network or to stop/lower the amount of users/programs connecting to it, in case they are causing problems?
(ideally, a drobo should be able to to its thing, while providing access to data, eg rebuild/expansions etc, access etc, but it’s also understandable that lots of “users” will make it slow down, but maybe just maybe in your case, theres something about all those “users” that are causing it to not work properly, so might be a good test to see if it helps?
it might help pinpoint some areas causing problems…[hr]
also… from your first post… you mention you had the phenomenon several times in the past too.
what was the outcome of those phenomenons (a tricky word to type quickly)
(or for south africans phiniminin) just kidding to lighten the mood…
but seriously, what was the outcome in the past… did the drobo correct itself after more hours/days, or did you have to start from scratch and reformat etc…
if it simply needed more time, and a restart or something, then maybe that solution will work again for you this time but i need more info
Hi Paul I have a mini most disapointing thing I ever brought, it’ll work fine for a few days if you carefully shut everything down each time. But every now and then it refuses to boot and your looking at an hour or more of switching things on and off before it miraculously decides to boot and you get the two green lights. I haven’t called support or spent much time on the forums but here’s the thing. It an EXTERNAL HARD DRIVE designed for photographers not a raspbery pi or a new programming language or an ESB. It shouldn’t need a forum it should just bloody work. Drobo needs to spend some time on testing these things on every platform they claim to support. It needs to just work like a TV or a Dishwasher right ootb first time or it’ll just be a geeks plaything for ever. I do a lot of programming photography web design, I work with a lot of enterprise software. But this thing is just way way harder than it should be, to get and keep working. I have two nas’s, and 4 external HDD collected over the last ten years and none of them do anything but work as soon as they’re plugged in.
hi xml, sorry just saw this now,
from an end-user perspective, you are right that and end user should just expect things to work and to keep working.
the unfortunate thing, is that a lot of items can be affected by different things, and do not always work how an end user would like.
eg i bought a toaster, and want it to always make toast,
but if i buy thick sliced bread of one make, and put my toaster on setting 2, it is fine
but if i use a different thick sliced bread from another make, setting 2 undercooks it
or burns it.
and if i add a slice of thick bread which has seedy bits and wholegrain nonsense ( i mean goodness), often burnt bits of seed fall off into the toaster and make a mess which requires some extra cleaning/maintenance.
(this is all true from my making a toast real example) but yes sometimes the analogies can apply (unfortunately) to other devices.
your suggestion to increase testing etc is perfectly fine