Hi Luke,
You have to use unify over afr to achieve this.
As an example consider a setup having 3 nodes n1, n2 and n3. Let the
requirement be all the .avi files to be replicated on 2 nodes (say n1 and
n2) and all the text files to be replicated on all the three nodes.
glusterfs configuration would be,
volume afr-for-avi
type cluster/afr
subvolumes n1 n2
#options follow
.
.
.
end-volume
volume afr-for-txt
type cluster/afr
subvolumes n1 n2 n3
#options follow
.
.
.
end-volume
volume unify
type cluster/unify
subvolumes afr-for-avi afr-for-txt
option scheduler switch
option switch.case *.avi:afr-for-avi;*.txt:afr-for-txt
#other options follow
.
.
.
end-volume
regards,
On Tue, Aug 5, 2008 at 3:51 AM, Luke McGregor <luke@xxxxxxxxxxxxxxx> wrote:
> Hi there
>
> On this site <http://gluster-tmp.enix.org/docs/index.php/GlusterFS_FAQ> it
> says that the switch scheduler can be used for file based regular
> expression
> redundancy. How exactly does this work and how does glusterfs keep the two
> copies consistent?
>
> Thanks
> Luke
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel@xxxxxxxxxx
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>
--
Raghavendra G
A centipede was happy quite, until a toad in fun,
Said, "Prey, which leg comes after which?",
This raised his doubts to such a pitch,
He fell flat into the ditch,
Not knowing how to run.
-Anonymous
