| 1 |
Boyd Stephen Smith Jr. wrote: |
| 2 |
> RAID-3?/5/6 can self-repair like this, but the checksumming is done at the |
| 3 |
> stripe, rather than inode level. |
| 4 |
|
| 5 |
AFAIK, RAID-5 doesn't self-heal except for the specific case where a bad |
| 6 |
block is detected by the hardware, so the RAID driver knows which drive |
| 7 |
has the bad stripe. If the parity is just inconsistent, there is no way |
| 8 |
of knowing which drive's stripe should be reconstructed. |
| 9 |
|
| 10 |
RAID-6 OTOH should be able to do that. |
| 11 |
|
| 12 |
> Surely, a filesystem should not shy |
| 13 |
> away from sanity checks that can be done with little overhead besides CPU |
| 14 |
> time, but adding a checksum to each block might be a little overkill. |
| 15 |
|
| 16 |
As long as performance is OK, I am willing to sacrifice the space for |
| 17 |
the per-block checksum. |
| 18 |
|
| 19 |
BTW, 10 drives? Nice setup! |
| 20 |
|
| 21 |
-- Remy |
Archives