| 1 |
On 2013-09-14 4:50 AM, Grant <emailgrant@×××××.com> wrote: |
| 2 |
> http://blog.open-e.com/why-a-hot-spare-hard-disk-is-a-bad-idea/ |
| 3 |
> |
| 4 |
> "Based on our long years of experience we have learned that during a |
| 5 |
> RAID rebuild the probability of an additional drive failure is quite |
| 6 |
> high – a rebuild is stressful on the existing drives." |
| 7 |
|
| 8 |
This is NOT true on a RAID 10... a rebuild is only stressful on the |
| 9 |
other drive in the mirrored pair, not the other drives. |
| 10 |
|
| 11 |
But, it is true for that one drive. |
| 12 |
|
| 13 |
That said, it would be nice is the auto rebuild could be scripted such |
| 14 |
that a backup could be triggered and the auto-rebuild queued until the |
| 15 |
backup was complete. |
| 16 |
|
| 17 |
But, here is the problem there... a backup will stress the drive almost |
| 18 |
as much as the rebuild, because all the rebuild does is read/copy the |
| 19 |
contents of the one drive to the other one (ie, it re-mirrors). |
| 20 |
|
| 21 |
> Instead, how about a 6-drive RAID 10 array with no hot spare? My |
| 22 |
> guess is this would mean much greater fault-tolerance both overall and |
| 23 |
> during the rebuild process (once a new drive is swapped in). That |
| 24 |
> would mean not only potentially increased uptime but decreased |
| 25 |
> monitoring responsibility. |
| 26 |
|
| 27 |
I would still prefer a hot spare to not... in the real world, it has |
| 28 |
saved me exactly 3 out of 3 times... |
Archives