Drive #2 was marked red this morning, and the dashboard
reported that the drive had failed.
I replaced it with a new drive, a WD 6 TB drive.
Dashboard then said the drobo was rebuilding, lights on the
box began flashing yellow-green, and estimated the rebuild
time at 51 hours.
At some point the dashboard lost the drobo. It can’t find it
any more.
The file system is still there and mounted … it behaves a little
strangely(*), but it seems to work. I’m not planning on using it much
until/unless this rebuild completes.
The host runs windows 7. I restarted the drobo dashboard
service but that had no effect.
Should I be concerned?
(*) I get occasional error messages that files no longer exist,
but then they open. Folders are impossible to delete.
This morning the array was in offline mode. Scary, but on power
cycle it came back up. It is not in rebuild mode any more, so
I can only guess that completed satisfactorily (much sooner than
it had forecast). It reports the new drive as healthy.
sometimes during a rebuild it is possible that dashboard (and/or the computer) can temporarily lose access to the unit. (i am not 100% as to why, and it may just be that in some operations the best way to stabalise that process could be to halt all access until a safer point in time, though usually the units keep working in terms of access when swapping out failed drives etc). in your case though i think you did the best thing which is to try to avoid using the drobo unless needed during a rebuild, as accessing or changing data during one can also slow down the overal usage and rebuild process.
one thing to keep in mind though, is that the estimate values can fluctuate (a bit like windows explorer) though a good value is to try to estimate how long it will take, by using approximately 1day per 1TB of data that you have on the drobo.
if you can remember how much data you had at the time, would that estimate correlate at all?