But this setup will be vulnerable to race conditions, for example
if server1 and server2 try to create file with same name in the
same path.
On 3/4/07, Krishna Srinivas <krishna@xxxxxxxxxxxxx> wrote:
Hi Gerry,
If there are four machines: server1 server2 server3 server4
server1 server spec file:
volume brick
type storage/posix
option directory /var/www
end-volume
### Add network serving capabilities to the "brick" volume
volume server
type protocol/server
option transport-type tcp/server
option listen-port 6996
subvolumes brick
option auth.ip.brick.allow *
end-volume
server1 client spec file will be like this:
volume client1
type storage/posix
option directory /var/www
end-volume
volume client2
type protocol/client
option transport-type tcp/client
option remote-host server2
option remote-port 6996
option remote-subvolume brick
end-volume
volume client3
type protocol/client
option transport-type tcp/client
option remote-host server3
option remote-port 6996
option remote-subvolume brick
end-volume
volume client4
type protocol/client
option transport-type tcp/client
option remote-host server4
option remote-port 6996
option remote-subvolume brick
end-volume
volume afr
type cluster/afr
subvolumes client1 client2 client3 client4
option replicate *:4
end-volume
here all read() operations happen on client1 as it is mentioned
first in the list of subvolumes. so it serves your purpose.
Let us know if you need more help.
On 3/4/07, Gerry Reno <greno@xxxxxxxxxxx> wrote:
> Krishna Srinivas wrote:
> > Hi Gerry,
> >
> > Could not really understand the requirements for your setup.
> >
> > Have you gone through the documentation?
> > http://www.gluster.org/docs/index.php/GlusterFS_User_Guide#Automatic_File_Replication_Translator
> >
> >
> Hi Krishna,
> Yes, this is the document I'm using. Let me describe in plain english
> what I need. I want to be able to setup a webserver farm using a
> clustered filesystem. In this way /var/www will be the same exact copy
> on all nodes. What is written to /var/www on any node should be
> replicated to the local storage brick on all the other nodes. This will
> simplify synchronizing all the webserver files and provide good
> performance because the webserver files will be local to each node.
> Can AFR do this? Because in reality wouldn't all the nodes be both
> master nodes and mirror nodes. If not AFR, then how can I do this?
>
> Thanks,
> Gerry
>
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel@xxxxxxxxxx
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>
