Need help determining which drive is going bad...

I’ve got a DroboElite running with Firmware version 2.0.4 [3.42.46950]

For a second time, I’ve gotten a warning that the unit can no longer protect against a failed drive, and then a notice that the raid is rebuilding. The rebuilding process takes WEEKS to complete, all the while performance when accessing suffers greatly.

When I look to see which drive has failed, all are operational.

This leads me to believe that a drive may be temporarily failing – failing enough to trigger a temporary drop of a drive, only to re-add it to the array - and thus trigger the raid rebuild.

Is there any way to find out which drive is causing this trouble? I know there are some support logs I can download, but I think those are encoded or otherwise protected from prying eyes?

hi snakebyte,
there might be a way to see if you view the dashboard tab, and to mouseover or click on the drive bay slots ,depending on the dashboard version, and it might show you a warning, or healed message?

apart from that, the only safer way so far, seems to be to generate a diagnostics log and to create a ticket with support. (two points here are that if your drobo is old and out of support, there would be a cost involved, a bit like a pay-per-view 1 time ticket, which might be worth it in the long run, but also that ideally diagnostic logs should only be run when the unit is in a stable state, because it might cause the rebuild problems or to force it to restart from the start)

No love with the mouseover. As far as I know I’m running the latest version of the dashboard for this unit.

Yeah, I did a diagnostic report and noticed that there was a hiccup from the drobo. At least it didn’t screw things up. I looked in the zip file and some of the log files weren’t encoded, but none contained any smart diagnostics info.

I think at this point what I should do is power the unit down and, one by one, run smartctl -x on them to see what the smart data says.

Does anyone know if it’s safe to power this thing down during a rebuild? I would use the dashboard to “shutdown” the unit.

thanks for the info, (you might have tried already but just to check did you also try clicking on each bay slot?)

ah ok - please dont try any more log generations in the meantime. you may have just stalled the current rebuild and it is continuing (usually at a rate of about 1 day per 1TB of data that you have on it), in which case if the rebuild has already been running for quite a while, it might be worth letting the rebuild process continue, for at least the estimated amount of time, plus a but of contingency before trying a power off.

i think there is in built resilience, such as in case there is a power cut during the rebuild i would ideally try to let the rebuild continue though, rather than power off, to play safe if you can