Greetings,

I'm trying to implement the following:

Node1, Node2 - Actual Nodes
Orat1, Orat2    - Alias Names for the specific Instances/SIDs

Orat1 will equate and answer to Oracle SIDs Test1, Test2
Orat2 will equate and answer to Oracle SIDs Test3, Test4

In normal operation Node1 will answer to requests for Orat1 and have the
Test1 and Test2 Instances/SIDs running on it.

In normal operation Node2 will answer to requests for Orat2 and have the
Test3 and Test4 Instances/SIDs running on it.

In planning a failover scenario, if Node1 goes down, using CAA the Orat1
(SIDs Test1 and Test2) will failover to Node2.

I'd like to know how to failover the broadcast of the Cluster Alias
(Orat1) over to Node2 so that Node2 will answer to the Orat1 Alias for SIDs
Test1 and Test2 - and visa versa when Node2 goes down and fails things over
to Node1.

I have looked at the caa scripts and have implemented things within the
cluster - except for changing of the announcements of the cluster alias and
munging with the /etc/clu_alias.config files when the nodes are in/are out,
etc.

Any ideas??

Thanks, I apprecate it.

Randy

[ Post a follow-up to this message ]

Anno Domini 31-12-2003 3:40, RCD sprak aldus:
quote:

>Greetings,
>
>    I'm trying to implement the following:
>
>    Node1, Node2 - Actual Nodes
>    Orat1, Orat2    - Alias Names for the specific Instances/SIDs
>
>    Orat1 will equate and answer to Oracle SIDs Test1, Test2
>    Orat2 will equate and answer to Oracle SIDs Test3, Test4
>
>    In normal operation Node1 will answer to requests for Orat1 and have th
e
>Test1 and Test2 Instances/SIDs running on it.
>
>    In normal operation Node2 will answer to requests for Orat2 and have th
e
>Test3 and Test4 Instances/SIDs running on it.
>
>    In planning a failover scenario, if Node1 goes down, using CAA the Orat
1
>(SIDs Test1 and Test2) will failover to Node2.
>
>    I'd like to know how to failover the broadcast of the Cluster Alias
>(Orat1) over to Node2 so that Node2 will answer to the Orat1 Alias for SIDs
>Test1 and Test2 - and visa versa when Node2 goes down and fails things over
>to Node1.
>
>    I have looked at the caa scripts and have implemented things within the
>cluster - except for changing of the announcements of the cluster alias and
>munging with the /etc/clu_alias.config files when the nodes are in/are out,
>etc.
>
>    Any ideas??
>
>Thanks, I apprecate it.
>
>Randy
>
>
>
>

http://h30097.www3.hp.com/docs/best...RA_SS/TITLE.HTM

You should give each SID its own alias.

--
-- "It's kind of...........a bummer." -- Ellen Feiss, student, 2002 --

[ Post a follow-up to this message ]