Hi
I wonder that about the pNFS's high availability.
The pNFS will stripe a file on each DSs, so this file will be currupted when one
DS has crushed, will not ?
On Tue, May 25, 2010 at 3:41 AM, redshield88888
<[email protected]> wrote:
> Hi
>
> I wonder that about the pNFS's high availability.
> The pNFS will stripe a file on each DSs, so this file will be currupted when one
> DS has crushed, will not ?
spNFS is an exercise in prototyping, useful for testing, and good for
demonstrating the power of the protocol to increase bandwidth. It is
not and has never claimed to be a production file system. There is
no 'fsck'. There is no heart-beat.. There is no stateid recovery, etc.
-->Andy
>
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to [email protected]
> More majordomo info at ?http://vger.kernel.org/majordomo-info.html
>
On 05/25/2010 10:41 AM, redshield88888 wrote:
> Hi
>
> I wonder that about the pNFS's high availability.
> The pNFS will stripe a file on each DSs, so this file will be currupted when one
> DS has crushed, will not ?
>
pNFS over objects can do that, and striping over mirrors, and more.
You need to use: pnfs-client with objlayoutdriver on client, exofs on MDS, and osd-targets
on disk-servers.
Cheers Boaz