the linux kernel stripes swap writes across multiple swap partitions of the
same priority. striping (effectively raid0) provides NO reliability in case of
disk failure. this is what the original poster was asking about.
to aswer the original question, i have twice tried to make swap on raid1,
which can be sucessfully done. the system swaps to the raid just fine.
HOWEVER- i pull the scsi data cable or power cable and the system will hang
cold on the next swap write.
i am using tekram dc390f (ncr53c875) scsi controller, and older 9gig
full-height seagate barracudas.
i have not had a chance to wrap my mind around the problem in question, and
the linux buff/cache code has a bit of a learning curve, but from practical
experience, i do not trust swap on raid 1, and wont use it. esp. since it
seems to have little to do with re-construction, and more to do with the death
of a disk, the very thing we want to prevent...
allan noah
"Jason A. Diegmueller" <[EMAIL PROTECTED]> said:
> : There has been much talk of problems with swapping on raid.
> :
> : As I understand it swapping on raid is fine unless the array
> : is reconstructing.
> [snip]
>
> I was under the impression that the current Linux
> swapping code automatically distributed (RAID-0
> style) across all available swap partitions.
>
> Is this not the case?
>
