User guide

147MAXDATA SR1202 M1 – StorView
®
RAID User Guide
Symptom Reason Solution
Inadvertently pulled the
incorrect drive from the
enclosure and the array
is dead.
Possible incorrect
drive identification and
removal.
If by mistake you remove a working drive
member instead of the failed drive, this can
cause the array to fail. In most cases you can
simply re-insert that drive that was incor-
rectly removed and the array will return to
the same state it was in prior to removing the
drive.
For RAID 5/50 arrays, a drive failure will put
the array in a critical state, if a hot spare was
available the array should go into a rebuild
mode. If you inadvertently remove one of the
known good drives that is in the process of
rebuilding, the rebuild operation will stop.
Once you re-insert the incorrectly removed
drive the array will return to the critical state
and the rebuild process will start again. If you
did not have a hot spare assigned, the array
will be in a critical state. If you inadvertently
remove a known good drive instead of the
failed drive the array will change to a failed
array state. Re-inserting that inadvertently
removed drive will put the array back into a
critical state. Replacing the failed drive will
cause the array to begin a rebuild operation
provided that you assign it as a hot spare or,
if the Auto Hot Spare option was enabled the
rebuild will begin automatically as the new
replacement drive is installed.
For RAID 0 arrays, if you inadvertently
remove a known good drive, the array will
become dead. Once you re-insert the incor-
rectly removed drive the array will return to
its working state.
For RAID 1/10 arrays, if you inadvertently
remove a known good drive, the array will
become failed. Once you re-insert the incor-
rectly removed drive the array will return to
its previous state. If the array was critical,
you can then replace the known failed drive
with a working drive and assign it as a hot
spare and the array will begin rebuilding.
NOTE: For all arrays removing a drive
as described above will cause all current
processing I/O from the controller to stop.
Any I/O in progress may have been lost or
cause a corrupt file. Be sure to verify all data
stored during this type of incidence to ensure
data reliability.