RE: RAC One Node - which server does instance start on?

  • From: "Shanmugam, Satheesh B" <satheesh.b.shanmugam@xxxxxxxxxxxx>
  • To: "jkstill@xxxxxxxxx" <jkstill@xxxxxxxxx>, Oracle-L Freelists <oracle-l@xxxxxxxxxxxxx>
  • Date: Fri, 30 Nov 2012 18:22:37 +0000

Did some poc on this sometime back. Managed to document some of my finding. 
Hope this gives you some info.

For installing RACONE we need to install:
1.      Grid infrastructure
2.      RAC software in all the nodes
3.      While creating DB using DBCA select RAC One node option. DBCA can be 
used only above In version we will have to use command line 
utility provided by oracle.  

Some of the issues found:
1.      This feature works best in and above. In version 
dataguard is not supported. Administration of RACONE in will be using 
command line tool. These tools are not user friendly. There is mandatory 
patch(#9004119) which needs to be applied to
2.      OEM 11.1 doesnât support RACONE.  It doesnât understand the failover or 
relocate option. 
3.      OEM 11.1 doesnât recognize the racone db automatically. We will have to 
manually add the entries to OEM 11.1. Any failover or relocate we will have to 
do it manually. 
4.      Configuration type in dbca should be âAdmin-Managedâ and not 
5.      When the inst_1 is running via racone, you can actually open the inst_2 
in node 2. Technically we can have both the instance running. Though Inst_2 
will use different service name than that of inst_1. This will confuse the DBA 
whether the itâs racone or not. But one good thing is all the racone instance 
will end with underscore and a number.

Crashing instance/node (failover):
                 When the node goes down or kill the pmon (kill -9 pmon), then 
cluster first tries to restart the instance in same node. If it is 
unsuccessful, it will start the instance in other node.  In the other node it 
will start with same name. Say If the instance name is inst_1, when it failover 
to node 2 it will be still inst_1.
                If you shutdown the db using shutdown abort/immediate, then 
cluster DOESNâT start the db automatically.
               The user sessions which are connected during failover will be 
disconnected. Users needs to open a new session.

               If you are relocating instance inst_1 from node 1 to node 2, it 
will create init.ora automatically in node 2. In node 2 it will start with 
different instance name, say inst_2.  
                The user sessions which are connected during switchover will 
remain connected. But any query which is running during the relocate phase will 
fail. User has to reissue the query.

Post switchover, it uses different instance name. In failover it retains the 
same instance name.

                You can convert the db from RACONE to RAC or vice versa using 
simple command. 

Some important commands,

srvctl relocate database -d inst_1 -n node2 âv ï inst_1 relocated to node 2. 

srvctl config database âd <db_unique_name> ï Using this command you can figure 
out whether the db is RACONE or not.

srvctl convert database âd <db_unique_name> -c RAC/RACONE ï Convert the db to 
rac or racone


Satheesh Shanmugam

-----Original Message-----
From: oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx] On 
Behalf Of Jared Still
Sent: Thursday, November 29, 2012 11:03 PM
To: Oracle-L Freelists
Subject: RAC One Node - which server does instance start on?

Yesterday I had my first experience with RAC One Node, and there is one feature 
of it that is somewhat confusing.
We installed  RAC One Node on an ODA, and all went smoothly.

When we started the database it appears the default node is the second node, as 
the database always starts on the second node.

We did use 'srvctl relocate database -d db -n new_node'  to relocate the 
instance to node A, and that works as expected.

So, we restarted the database, using srvctl from node A.

Again the instance started on Node B.
 Googling and Oracle support haven't been helpful for this so far.
 BTW, using 'start instance' with -n node or -i instance is not supported with 
RAC One Node.

Oracle strongly recommends that database connections are made through a 
service, and I suspect that is due to not knowing which node the database may 
be running on, All indications so far are that this is the way it works 
normally, but there does not seem to be a lot of experience with RAC One Node 
in the community.

Any ideas?

Jared Still
Certifiable Oracle DBA and Part Time Perl Evangelist Oracle Blog: Home Page:


This communication is for informational purposes only. It is not
intended as an offer or solicitation for the purchase or sale of
any financial instrument or as an official confirmation of any
transaction. All market prices, data and other information are not
warranted as to completeness or accuracy and are subject to change
without notice. Any comments or statements made herein do not
necessarily reflect those of JPMorgan Chase & Co., its subsidiaries
and affiliates.

This transmission may contain information that is privileged,
confidential, legally privileged, and/or exempt from disclosure
under applicable law. If you are not the intended recipient, you
are hereby notified that any disclosure, copying, distribution, or
use of the information contained herein (including any reliance
thereon) is STRICTLY PROHIBITED. Although this transmission and any
attachments are believed to be free of any virus or other defect
that might affect any computer system into which it is received and
opened, it is the responsibility of the recipient to ensure that it
is virus free and no responsibility is accepted by JPMorgan Chase &
Co., its subsidiaries and affiliates, as applicable, for any loss
or damage arising in any way from its use. If you received this
transmission in error, please immediately contact the sender and
destroy the material in its entirety, whether in electronic or hard
copy format. Thank you.

Please refer to for
disclosures relating to European legal entities.--

Other related posts: