[Gluster-users] Is unify&replicate configuration impossible?
이재근
overaction00 at gmail.com
Wed Sep 2 06:02:00 UTC 2009
I try to make unify & replicate gluster file system.
I have 3 machine for 3server, 1client( 1 machine both server and client)
machine 1 : server , client machine 2 : server machine 3 : server
first, machine 1, 2 are configured to unify (I'm planning to use the
algorithm nufa)
then I conbined it with machine 3 to provide AFR by configuring replicate.
but it can't operate properly
here is my volume file.
________________________________
glusterfsd.vol
volume strg
type storage/posix
option directory /data/export
end-volume
volume strglocks
type features/locks
subvolumes strg
end-volume
volume expdir
type performance/io-threads
option thread-count 8
subvolumes strglocks
end-volume
volume nspace
type storage/posix
option directory /data/namespace
end-volume
volume svol
type protocol/server
option transport-type tcp
option auth.addr.expdir.allow *
option auth.addr.nspace.allow *
subvolumes expdir nspace
end-volume
________________________________
glusterfs.vol
volume server1
type protocol/client
option transport-type tcp
option remote-host 147.6.3.41
option remote-subvolume expdir
end-volume
volume server2
type protocol/client
option transport-type tcp
option remote-host 147.6.3.46
option remote-subvolume expdir
end-volume
volume server3
type protocol/client
option transport-type tcp
option remote-host 147.6.3.47
option remote-subvolume expdir
end-volume
volume client1-namespace
type protocol/client
option transport-type tcp
option remote-host 147.6.3.46
option remote-subvolume nspace
end-volume
volume clients
type cluster/unify
option scheduler nufa
option nufa.local-volume-name server1
option namespace client1-namespace
subvolumes server1 server2
end-volume
volume replicate
type cluster/replicate
subvolumes clients server3
end-volume
volume writebehind
type performance/write-behind
subvolumes replicate
end-volume
It can mount glusterfs but, can not copy, remove, etc.
Is this configuration possible?
More information about the Gluster-users
mailing list