Upgrade 10.2.0.2 RAC database to 10.2.0.4

Hi,
I have two RAC databases(say dev and qa) having the same ORACLE_HOME. The oracle version is 10.2.0.2.
Now I want to upgrade only the Dev database to 10.2.0.4.
For this upgarde process, I am planning to proceed as below
1) Install 10.2.0.2 binaries in new Oracle Home
2) Apply the patch for 10.2.0.4 on 10.2.0.2 new home to upgrade the binaries to 10.2.0.4
3) Change the reference of the oracle Database i.e. Dev only to the new oracle home using srvctl
4) Then start the database in startup upgrade and run catpatch.sql .
I am not sure whether it is the correct way to proceed.
Please let me know is it the correct way to follow for upgradation, if not please give the excat possible way to upgrade to 10.2.0.4
Thanks in advance,
Mahi

If you want a step by step description of exactly what commands to execute, you would need to post more information than you would ever divulge. Then someone would surely write the commands for you. But then, they would also have to charge you because that's what consultants are for.
Have you read the README that comes along with the upgrade? Have you read the installation manuals for your as of yet undisclosed platform?
http://download.oracle.com/docs/cd/B19306_01/relnotes.102/b14265/toc.htm#NTDNX110

Similar Messages

In-place upgrade of Enterprise Manager for RAC databases is not supported

upgrading 2 node RAC from 10.2.0.1 to 10.2.0.3 on Windows 2003 64 bit. I got the following:
In-place upgrade of Enterprise Manager for RAC databases is not supported in this release
Now I cannot log in to the database control as sys with sysdba role . I get "invalid username/password" error
I can login as a dba user
I also get a java error after I log in, some thing about class not found, but then I'm in and I can do everything that I try, although I haven't tried scheduling a backup yet
Can I upgrade EM to 10.2.0.3?
If so, how?

On the patchset readme...
11.3 Upgrade of Oracle Enterprise Manager Not Supported
When you start Oracle Database Upgrade Assistant to upgrade 10.2.0.1 database to 10.2.0.3 patch set, the following error occurs:
Inplace upgrade of Enterprise Manager is not supported for RAC databases is not supported in this release
Workaround:
Ignore warning.
If you drop and recreate the EM repository?
1). emca -deconfig dbcontrol db repos drop
2). emca -config dbcontrol db -repos create
Be careful, doing this put the database on QUIESCE mode.
Regards,
Rodrigo Mufalani

Upgrade of RAC database in 10g or 11g

Hi,
Can we upgrade RAC database without downtime like we do patch?
Regards

user602441 wrote:
Hi,
Can we upgrade RAC database without downtime like we do patch?
RegardsHi,
The Oracle Clusterware software always fully supports rolling upgrades, while the ASM software is rolling upgradeable at version 11.1.0.6 and beyond.
Rolling upgrade: we mean upgrading software (Oracle Database, Oracle Clusterware, ASM or the OS itself) while the cluster is operational by shutting down a node, upgrading the software on that node, and then reintegrating it into the cluster, and so forth one node at a time until all the nodes in the cluster are at the new software level.
For the Oracle Database software (RAC), it is possible only for certain single patches that are marked as rolling upgrade compatible. Most Bundle patches and Critical Patch Updates (CPU) are rolling upgradeable. Patchsets and DB version (10g to 11g) changes are not supported in a rolling fashion, one reason that this may be impossible is that across major releases, there may be incompatible versions of the system tablespace, for example. To upgrade these in a rolling fashion one will need to use a logical standby with Oracle Database 10g or 11g.
Regards,
Levi Pereira

Upgrading clusterware and RAC database including ASM

Hi All,
I have a 3 node RAC database with ASM running on RHEL4(32bit). current version is 10.2.0.1. I want to upgrade clusterware, database and ASM to 10.2.0.4.
Please let me know the patch number for this upgradation.
At the time of googling i found that 1st we have to upgrade 10.2.0.1 to 10.2.0.3 then we can upgrade it to 10.2.0.4. Is it true?
Thanks

You are absolutely correct. Database should be upgraded at last. I may be wrong regarding upgrading of ASM. but what my thought is saying it should be upgraded or it may upgrade when we will upgrade our database. So please think little bit on it. ASM has its own instance. we are upgrading clusterware and database. why we are not upgrading our ASM. Then what should be the method to upgrade ASM.
Please correct me in i am wrong.
With Regards,
snp

Restoring Back To 10g After A Failed 11g Upgrade Of a RAC Database

I'm testing this out on a small test RAC database. I successfully upgraded it from 10.2.0.4 to 11.2.0.2 but wanted to test the scenario of having to go back to 10g if the upgrade really hosed up. The first recovery attempt seemed to be successful but after bringing the DB down with srvctl, it failed on the next startup saying it needed to be started in upgrade mode. Something from 11g was still in place or the fact that I was trying to restart a 10g database managed by 11g clusterware was the issue. I tried starting the DB from both 10g and 11g environments and got the same result. Even starting each instance individually got the same result.
In all that I tried, I got the usual incarnation and "until time before reset time" messages. I've been doing this all through RMAN without EM or Grid Ctl. As usual, any docs I found have had just a little information and I have to piece my own instructions together from all of them, not knowing for sure it all steps would apply in my situation.
Can anybody point me to a good doc or other resource that might help me out? Many thanks!

Now I have a different issue with apparently the same problem. I successfully did the restore/recover as before but the thing pukes when I open resetlogs at the end. Something, somewhere is still pointing to 11g but I have no idea where or what has changed since the last time I did this. Maybe I've messed things up by doing this multiple times. Here are my RMAN commands:
RMAN> connect target
RMAN> startup force nomount;
RMAN> RESTORE SPFILE TO '+DATA/jimg/spfilejimg.ora' from '/local/oracle/10.2.0/db_1/dbs/c-2526333028-20110915-01';
RMAN> shutdown immediate;
(in a different session, from command line)
% mv /local/oracle/10.2.0/db_1/dbs/initJIMG1.ora.bak /local/oracle/10.2.0/db_1/dbs/initJIMG1.ora
RMAN> startup force nomount pfile='/local/oracle/10.2.0/db_1/dbs/initJIMG1.ora';
RMAN> restore controlfile from '/local/oracle/10.2.0/db_1/dbs/c-2526333028-20110915-01';
RMAN> alter database mount;
RMAN>
run {
restore database;
recover database;
RMAN> alter database open resetlogs;
The errors I get after resetlogs are:
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of alter db command at 09/16/2011 09:24:50
ORA-01092: ORACLE instance terminated. Disconnection forced
ORA-00704: bootstrap process failure
ORA-39700: database must be opened with UPGRADE option
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
ORA-03114: not connected to ORACLE
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of alter db command at 09/16/2011 09:24:50
ORA-01092: ORACLE instance terminated. Disconnection forced
ORA-00704: bootstrap process failure
ORA-39700: database must be opened with UPGRADE option
This is being done on a small, expendable database for practice. I think I've learned enough about what NOT to do to keep me from getting in this jam to begin with. Any thoughts?

Migration of 10g RAC database to new sever which includes 11g upgrade

Hello all,
I have requirement here to migrate 10g RAC database from one server to another server as part of migration i want to perform upgrade to 11g as well.
To Help me out.Please post your inputs or way to perform the task.
Precautions to be taken while doing the activity.
Please find the below for more info
Old Box:
Current environment
RAC -2node
ASM
Database Version 10.2.02
Os:HP-UX 11.21
New Box:
New environment should be
RAC -2node
ASM
Database Version 11.2
Os:HP-UX 11.31
Thanks for the help in advance
Anand

Pl do not post duplicate threads - Migration of 10g RAC database to new sever which includes 11g upgrade

Help needed in deleting nodes from RAC database

Our DB is 10g RAC database and the servers are Window 2003 server. Initially the database was configured to have 4 nodes. For some reason we stopped the instances in the first two nodes and the current database is running on node3 and node4. I queried the v$thread view and it shows 3 records. Thread 2 is closed and disabled. Thread 3 is open and public and thread 4 is open and private. Now we need to disconnect nodes 1 and 2 from RAC db and cluster ware (We use Oracle cluster ware) and plan to use those two servers for other purposes. Although I read through the Oracle doc regarding deleting node from RAC “11 Adding and Deleting Nodes and Instances on Windows-Based Systems” and wrote down the steps we need to take but I am still not comfortable in doing it since it is production env and we don’t have any dev env to practice those steps. So I would like to borrow your experiences and your lessons learned while you were doing those. Please share your thoughts and insights with us.
Thank you so much for your help,
Shirley

what's your full version? I can warn about specific issues in 10.1.0.3 and 10.2.0.1 - for example, in some cases (depending on how the listener was configured), it will be impossible to delete the listener for the deleted node. Known bug.
To avoid many many known bugs, you may want to upgrade to at least 10.2.0.2 before removing a node (from my experience, this is the first stable RAC version).
In any case, deleting a node is a rather delicate process. I really really recommend practicing. Take any pc/laptop, install VMWARE, define two virtual machines, install RAC and remove one node. It will take you an extra day or two, and could save your production.

How to install Oracle 11.2.0.4.0 Rac Database on Grid Infrastructure 12.1.0.2.0 (Windows 2012 R2)

Hi all.
This is the scenario:
I have a 2 node cluster Microsoft Windows 2012 R2 x86_64, the nodes are part of domain.
Installed Oracle 12cR1 (12.1.0.2.0) as follows:
Installation done using domain administrator named "installoracle" that has been EXPLICIT promoted to local administrator in both nodes.
One domain user named "oracle12c" with no administrator privileges neither in the domain or nodes.
Installation, after a lot of customization in the OS of the nodes, has gone ok, perfect indeed, not even a warning during runinstaller execution.
All services and resources are up & running, I have rebooted nodes and all comes back ok in place. In windows services I can see some services started with "oracle12c" user, and all windows groups: DBA, SYSKM, etc, has been created.
Now I have to install Oracle RAC database in 11gR2 version (11.2.0.4.0) to accomplish with application requirements.
So, I execute runInstaller (setup.exe) logged as administrator via domain administrator "installoracle" (the one that was able to do the smooth 12c GI installation) and the gui comes with out problems, but at 4th screen the one in which you have to choose what kind of installation you want to do (Single, RAC or RoN) I select RAC option, the gui returns error: crs is not running in the local node...
Finally, in order to deliver some database 11.2.0.4.0 for my customer, I try to do installation of SW for Single Node, I did using "installoracle". The sw has been installed in one of the nodes. Then I have created a database and I have stored datafiles in ASM (this ASM is part of 12c clusterware) and also register the 11gR2 single instance database in the Grid Infrastructure listener
So, whats next??? Why is this behavior??? I did similar installations in Linux/Solaris with no issue (well as everyone can imagine, always there are some issues, but when solved you got GI in 12c and databases in the release that you want 12c, 11g, 10g...)
Any clue?
Thanks in advance!!
Best Regards!!

To go back to old version (without changing compatible parameter), you can perform downgrade.
Instructions are available at
How to Downgrade from Database 11.2 to Previous Release (Includes 11.2.0.4 - 11.2.0.1) (Doc ID 883335.1)
In this case you will having all the data file including the new one.
But remember database may not be in the same status as before (Before upgrade). it may have invalid objects of higher version after downgrade.
Downgrade is applicable only to successfully upgraded databases. In case upgrade faced some issue, do not try downgrade there.
Your second approach it not correct. Note that each datafile header will have db version. We can not open database in normal mode when header version and binary versions are different.

Recommended patches on oracle 10.2.0.4 RAC database and CRS

Hi,
This week I am planning to upgrade my oracle 10.2.0.2 RAC database oracle 10.2.0.4.
Also, I want to apply if any recommended patches we can apply on 10.2.0.4 database or on CRS 10.2.0.4
I searched in metalink, but in vain i didn't get any exact patches recommended on 10.2.0.4.
If you have any patch numbers that need to be applied on oracle 10.2.0.4 RAC database, please provide me.
Thanks,
Mahi

Check out MOS Doc 756671.1 - Oracle Recommended Patches -- Oracle Database
HTH
Srini

RAC database patchset 10.2.0.5 on windows

HI,
I have 2 RAC environment having 13 Production database with version 10.2.0.4 of database, clusterware and ASM on Windows server 2008 64 bit.
Now I need to apply patchset 10.2.0.5 on ONLY one database due to BUG. Instead of apply patchset on all RAC database and upgrade planning for only one affected database. I need to below clarification -
1. I am planning to apply patchset 10.2.0.5 on clusterware and ASM only.
2. Planning to install 10.2.0.5 on new oracle home.
3. Planning to upgrade one affected database from 10.2.0.4 to 10.2.0.5.
QUESTION HERE -
1. Will above plan work for one database only.
2. Will clusterware and ASM (10.2.0.5)will support both ORACLE_HOME (10.2.0.4 and 10.2.0.5) databases
3. How we can setup clusterware for this new ORACLE_HOME (10.2.0.5) DATABASEs.
Anyone have any plan or suggestion please share.

Hi,
Yes it is possible to have multiple ORACLE_HOMES inside a Oracle GRID infrastructure. To check the supported and compatibility My Oracle Support for this.
1. check compatibility of the components and make the design for your listener
2. Upgrade ASM en clusterware. to 10.2.0.5
3. Install new oracle software tree 10.2.0.5
4. upgrade one database to 10.2.0.5
keep in mind the Oracle Cluster Registry for this database has to change
Cheers,
Jos van den Oord
Blog : http://joordsblog.vandenoord.eu/
Company : http://www.transfer-solutions.com/
"Don't believe it, test it!"

While creating RAC database on IBM AIX 5.3 following errors occur.

Hi,
I am creating a Oracle database on following configuration IBM AIX 5.3 box:
OS: IBM AIX 5.3
DB: Fresh installed Oracle 9.2.0.1 upgraded to Oracle 9.2.0.7 (Patch)
While creating a RAC database on the above server i got following errors:
ORA:00603 : ORACLE SERVER SESSION TERMINATED BY FATAL ERROR
ORA-27504 : IPC error 380 OSD context
ORA-27300 : OS system dependent operation : if_not_found failed with status:0
ORA-27301 : OS failure mesage: error 0
ORA-27302 : failure occured at:skgxpbaddr9
Licence high water mark = 0
Can any 1 help me out.Are the above error related to OS or something is wrong with the OS cluster / RAC. I tried creating single instance database on either of the server which are in cluster, no success.
Also 1 more thing i would like to add is that iam able to create single instance database on server which are not in cluster.
Any help regarding the above error is appreciated.

This is a bug (for patchsets 9.2.0.6 and higher), see metalink:
UNABLE TO START RAC INSTANCE AFTER APPLYING PATCHSET 9.2.0.6 ON AIX
Note:300218.1
Werner

Restore RAC database from non-RAC source - EBS R12.1.1

We have restored (PIT) a RAC database from a non-RAC (but ASM) source and I was wondering if that was a reasonable thing to do.
The reason I ask is that I seem to remember that in the upgrade to R12.1.1 the upgrade driver had a different list of jobs to do when the target was RAC vs when it was non-RAC... so is it possible the EBS R12.1.1 code is different for RAC (meaning we dont have the right code base in the DB since we restored a non-RAC source to a RAC destination)?
Disclaimer - I am NOT a DBA .. just trying to identify some potential problems we are troubleshooting.
Thanks in advance!
EBS R12.1.1
OEL 5
11gR2

We have a RMAN back up of the non-RAC database that was restored and PIT recovered on the RAC database.Is it a full database backup?
The "code" base I am referring to is all the db stored objects for R12 EBS .. database compiled code.Restoring the RMAN backup on the target node is not enough. Have you run AutoConfig or postclone? Does your instance know that this is a a single instance and not a RAC one?
I am nervous that there may be alternate versions of the code to run in RAC vs non-RAC .. for instance I know that some of the ASCP jobs use DBMS PIPES and that can be tricky with RAC and Concurrent Manager.. And when we did the upgrade to R12, pointing it @ RAC generated more and different jobs than when it was not pointed @ RAC .. so SOMETHING was different according to the APPS ADPATCH/ADADMIN utilities.Please explain in details what is the purpose of doing this restore and what exactly you are trying to achieve. What cloning procedure you follow to refresh the target instance?
Thanks,
Hussein

Open 10g/11g RAC database in Non RAC Mode

Dear Friends,
How i open 10g and 11g RAC database in Non RAC Mode (open single instance in order to perform upgrade)
Later i will convert the upgraded DB to RAC DB
Regards,
DB

user564706 wrote:
Dear Friends,
How i open 10g and 11g RAC database in Non RAC Mode (open single instance in order to perform upgrade)
Later i will convert the upgraded DB to RAC DB
Regards,
DBEdit spfile and set cluster_database=false before starting database in upgrade mode(optionally you can unset other cluster parameters).
So you upgrade using only 1 instance of database.
Once the upgrade is finished start the other instance as well and set cluster_database=true
Also see preupgrade process - http://docs.oracle.com/cd/E11882_01/server.112/e23633/upgrade.htm#i1011979
Post upgrade -> http://docs.oracle.com/cd/E11882_01/server.112/e23633/afterup.htm#CEGCFJJC

Can I put ALL my applications into one RAC database

Dear all,
Is it possible to share all of my workloads between a single RAC database or would it be better to run more than one RAC database on the the same cluster if they have very different characteristics (e.g. OLTP vs DSS or sensitive vs non-sensitive data). Oracle seem to imply that they have cut down their number of physical databases drastically but have they been able to do that because the workload is very homongenous, i.e all ERP stuff? Should it be possible to accommodate all workloads by having different instance parameters or does that defeat the object of being able to fail over a service from one instance to another?
Views/experiences please.

Hi, generally it is possible to put everything into one RAC database, especially if they share the same data. So you do not have to transport data from one database into the other.
Furthermore the resource allocation can perfectly be addressed with the resource manager of the database together with services. The workload has not to be homogenous.
However it might make sense to connect the DSS users to different nodes in the cluster than the OLTP users (via. services).
On the other hand there might be non technical issues which would make seperate databases more attractive. Like Patching could mean downtime of all applications. Like certifications of software, which would not allow you to upgrade your database, like ERP needs 10.2.0.2 while DSS would be better off with 11g) and so on.
There is really no right or wrong.
Still to get the most of the benefits out of a cluster the fewer databases you have.

Metalink Document IDs for RAC database

Hello,
This is Khalid, Plz any body could give me metalink document id's for RAC database managing and monitoring, backup and recoveries etc.. plz any document id related to RAC databases.....
Thanks & Regards
Khalid

Clusterware References
Metalink Notes
259301.1      CRS and 10g RAC
     This note contains a useful awk script to improve the output of crs_stat -ls
436067.1      Windows CRS_STAT script to display long names correctly
309541.1      How to start/stop the 10g CRS Clusterware
263897.1      How to stop Cluster Ready Services (CRS)
298073.1      How to remove CRS auto start and restart for a RAC instance
295871.1      How to verify if CRS install is valid
316583.1      VIPCA fails complaining that interface is not public
341214.1      How to cleanup after a failed (or successful) Oracle Clusterware installation
280589.1      How to install Oracle 10g CRS on a cluster where one or more nodes are not to be configured with CRS immediately
357808.1      CRS Diagnostics
272331.1      CRS 10g Diagnostic Guide
330358.1      CRS 10g R2 Diagnostic Collection Guide
331168.1      Oracle Clusterware consolidated logging in 10gR2
342590.1      CRS logs not being written
357808.1      Diagnosability for CRS/EVM/RACG
459694.1      Procwatcher: Script to Monitor and Examine Oracle and CRS Processes
289690.1      Data Gathering for Troubleshooting RAC and CRS issues
265769.1      Troubleshooting CRS Reboots
240001.1      Troubleshooting CRS root.sh problems (10g RAC)
239989.1      10g RAC - Stopping Reboot Loops when CRS problems occur
294430.1      CSS Timeout Computation in 10g RAC
284752.1      10gRAC: Steps to Increase CSS Misscount, Reboottime and Disktimeout
462616.1      Reconfiguring the CSS disktimeout of 10gR2 Clusterware for proper LUN failover
293819.1      Placement of voting and OCR disk file in 10g RAC
317628.1      How to replace a corrupt OCR mirror file
452486.1      Moving OCR and Voting Disk to another location
399482.1      How to recreate OCR/Voting disk accidentally deleted
358620.1      How to recreate OCR/Voting disk in 10gR1/R2 RAC
279793.1      How to Restore a Lost Voting Disk in 10g
264847.1      How to Configure Virtual IPs for 10g RAC
283684.1      How to change interconnect/public interface IP subnet in a 10g cluster
276434.1      Modifying the VIP or VIP Hostname of an Oracle 10g Clusterware Node
294336.1      Changing the check interval for the Oracle 10g VIP
219361.1      Troubleshooting Instance Evictions (ORA-29740)
297498.1      Resolving Instance Evictions on Windows platforms
315125.1      What to check if the Cluster Synchronization Services daemon (OCSSD) does not start
270512.1      Adding a node to a 10g RAC Cluster
269320.1      Removing a node from a 10g RAC Cluster
338706.1      Cluster Ready Services (CRS) rolling upgrade
399031.1      Step-by-step installation of Oracle Clusterware one-off and bundle patches for Oracle 10g
401783.1      Changes in Oracle Clusterware after applying 10.2.0.3 Patchset
405820.1      Known Issues After Applying 10.2 CRS bundle patches
316817.1      Cluster Verification Utility (CLUVFY) FAQ
372358.1      Shared disk check with the Cluster Verification Utility
338924.1      CLUVFY Fails with error - could not find a suitable set of interfaces for VIPs
Bugs
5849200      CRS LOGS ARE NOT BEING WRITTEN
5137401      OPROCD LOGFILE IS CLEARED AFTER A REBOOT
     Fixed in Oracle 10.2.0.4+ and 11.1.0.6+
source: http://www.juliandyke.com/References/Clusterware.html
regards,
Rajeshkumar Govindarajan.
http://oracleinstance.blogspot.com

Upgrade 10.2.0.2 RAC database to 10.2.0.4

Similar Messages

Maybe you are looking for