|
A Nirvana Realm is an instance
of the server and a container for channels, queues and
P2P services. Each Nirvana Realm defines a namespace
of its own but it is possible to merge the namespaces
of multiple Realms into one large one.
While adding a Nirvana Realm into the namespace of
another using the Realm manager you need to know the
RNAME of the Realm you are adding in to your the
source Realm.
Note: Before adding a Realm to another Realm make sure
that the destination Realm has set permissions such
that the source Realm has permission to carry out operations.
Note: In order to add one Relm to another's namespace
neither Realm can have the same name. By default when
a Nirvana Realm is being installed it is called Nirvana.
To give the Realm a different name then choose the custom
install option during install and give it a different
name. Alternatively after installation the nserver.lax
file can be edited to change the name. This iis achieved
by editing the -DREALM property.
For alternative information on adding a Nirvana Realm
to another Realm's namespace please see the following:
Step 1: Right click on the Realm node
on the Nirvana namespace of your 'primary' realm manager.

Step 2: Choose the Add Realm option
and specify the RNAME of the Realm you wish to add to
the namespace. In addition specify a mountpoint so that
all of the resources of the Realm you add will become
available to sessions under the name /mountpoint.

In the dialog box above the a Realm
running on port 13002 on a host called wudan is added
by specifying the RNAME nsp://wudan:13002. Once the
Realm has been added it will appear in the original
namespace under the mountpoint of /bkup.

Sessions connected to the 'primary'
realm now hace access to two channels. These are :
- /FooChannel which is a local channel and;
- /bkup/FooChannel which is actually a channel on
another as specified by the RNAME above.
Once more than one Realm has been added to a namespace
channel joins can be added, and the namespace can also
be configured such that it can be used for automatic
HA failover. |