RAC OS upgrade

I'm looking for more information on the upgrade of the OS in a 4 node RAC cluster. Specifically, I need go RHEL v4.3 to v4.7 RHEL. Y at - he of the documentiaon on this process? I had no luck finding one. For minor updates, you can remove the node, update, and add on the cluster? Thanks for any help/feedback.

Hey Buddy,

I'm looking for more information on the upgrade of the OS in a 4 node RAC cluster.
Specifically, I need go RHEL v4.3 to v4.7 RHEL. Y at - he of the documentiaon on this process?
I had no luck finding one. For minor updates, you can remove the node, update.
and add to the bunch? Thanks for any help/feedback.

If you want to follow the system, you may need to upgrade from node to node, if you can keep the system down for maintanace stop the CRS in all the nodes.

Basically, if you want to upgrade the operating system with your system to the top, you will have to roll OS upgrade in a cluster (node in node upgrade)

The procedure would be something like that.

1 - Navigate to the node 1

1.1 - terminating the instance

srvctl stop instance d -i

1.2 - stop ASM in this node

srvctl stop asm - n

1.3 - stop the nodeapps

srvctl stop nodeapps - n

1, clusteware 3-stop (run as root user)

crsctl stop crs

2 upgrade/patch the OS

3 - create a link to the binary files for RDBMS

3, 1 - cd $RDBMS_HOME/bin

resynchronize all

4 - create a link to the binary files for ASM

3, 1 - cd $ASM_HOME/bin

resynchronize all

PS: CRS doesn't have to be reprinted when

5 - start of the cluster in this node (run as root)

crsctl start crs

6 - Check if all the resources are up and valid

7 - repeat the procedure in steps 1-6 for the rest of the nodes in this clusteware, by node.

The references you want are:

1-certification matrix:

"Oracle Clusterware and Real Application Clusters support mixed servers in the same cluster,
as long as the servers are running the same operating system (e.g., OS even binary).
You can create a cluster of machines of different speeds and size.
In addition, Oracle supports the rolling upgrade of the operating system in a cluster
When the two versions of the operating system are certified with Oracle database version that you are running.
Note that the mixed operating system versions are supported only for the duration of an upgrade (i.e., 1-2 hours)
-the cluster must never operate with mixed operating systems for an extended period.
Oracle Clusterware and Real Application Clusters do not heterogeneous platforms support
(for example, machines with different chip architectures) in the same cluster. »

2. Metalink notes: 444595.1 (is it necessary to recreate a link to following Oracle OS Upgrade?)

3 - Metalink note: 220970.1 (RAC: frequently asked Questions)
"Recreate a link it takes to CRS_HOME after the operating system upgrade?
Binary Oracle Clusterware cannot be relinked on shiphomes. So to answer your question, no, there is no need to recreate a link to the binary Oracle Clusterware. "

PS: REMEMBER TO VALIDATTING THE PROCESS OPENING AN SR FOR THAT! VERY IMPORTANT!

It will be useful,
Cerreia

Tags: Database

Similar Questions

  • Nothing special about RAC DB upgrades?

    The GI version: 11.2.0.4

    DB version: 10.2.0.5 and 11.2.0.4

    Platform: Oracle Linux 6.4

    Number of nodes: 3

    We have a DB 10.2.0.5 running in an GI 11.2. We want to upgrade this DB in a DB 11.2.

    We already have 11.2.0.4 RDBMS installation in all 3 nodes. That's what I intend to do, to update our data base RAC 10.2 to 11.2 CARS DB.

    Step 1. Run the utlu112i.sqlin "pre-upgrade information tool" g's existing 10 DB.

    Solve all of the problems mentioned in the output

    Step 2. Stop the db 10g

    Step 3. Copy the new file inti.ora with the new 11g setting in the ORACLE_HOME/dbs 11.2. Delete the old settings of 10g

    the value CLUSTER_DATABASE = FALSE

    Step 4. Set ORACLE_HOME as a new ORACLE_HOME 11.2

    Step 5. set the ORACLE_SID for 10g DB and start the instance only in node 1 in UPGRADE mode

    $ export ORACLE_HOME = < location of Oracle 11.2 >

    $ export PATH = $ORACLE_HOME/bin: $PATH

    $ export ORACLE_BASE = < Oracle_Base defined during installation 11.2 >

    SQL > start Update pfile=11g_R2HOME/dbs/init.ora

    Step6.run 11gR2_home/rdbms/admin/catupgrd.sql

    Step 7. After running catupgrd.sql, start the instances of Node2 and Node3.

    The peculiarity only I need to do is to put CLUSTER_DATABASE = FALSE. Is there anything else I need to do?

    I want to do it manually because I have more control over the upgrade process.

    What do you hope to gain from having more control? What would be the advantage?

    Remember that you do not have * control more. You have the complete control and with it, all the responsibilities. In order to better not forget a step that would have made DBUA.

    Furthermore, I've never used DBUA in Standalone or DBs CARS before. Need to learn

    Have you read the last line in my first reply to this topic? I hope that you will test your upgrade before you do in production. You can read more of my thoughts on the tests:

    Importance of testing

    Testbeds

    Because you will test this point, why not take this as an opportunity to learn how the DBUA is simple to do? Sound far, much easier than the manual upgrade.

    Are the steps mentioned in my first post enough?

    I already mentioned some of the things you forgot in my first answer. Here they are again:

    You will need to manually fix the configuration of your network files. You have to manually manage the spfile and password file to new homes. And you don't have any stage to tell the new directory of Oracle clusterware.

    None of these steps has appeared in your list. Other steps that I can quote are to update the oratab file, manually start all RAC instances, check your services are up and running, recompile invalid objects. In addition, I saw no mention of backups. It is wise to backup before and after the upgrade. The backup before the upgrade is desirable in case the upgrade is very bad. You can return to the old version. The upgrade after upgrade is complete, as expected, a failure occurs, you will not have to redo the upgrade after restoring a pre-upgrade backup.

    That being said, learn to use the DBUA. Let the tools provided by Oracle as a DBA easier. In addition to the economy you time, the DBUA decreases the risk of human error in the upgrade process. Of course, I would go in my bosses office after I missed a db upgrade to see my boss to know that if I used the right tool for the job, the upgrade would not be smooth.

    See you soon,.
    Brian

  • Plugin Flash races to come, has no sound, since the upgrade to Ubuntu, why?

    After update Ubuntu 11.04 to 12.04 (64-bit), embedded videos managed by the Flash plugin race advance to the download speed, totally ignoring the clock. They are also silent.

    Found the problem. It is an odd. Research in the application of the Volume control (not sure if it's one thing to Gnome or Xfce - desktop is Xfce), he showed that he was failing to try to play the audio stream through "Redwood HDMI Audio" due to a Radeon video card in the system. Turn off sub tab Configuration and having integrated Audio together the appropriate (in my case) audio analog stereo sound restorations that it fixes the problem Flash video--race to advance.

    What is crazy is that the video should come race due to an audio output doesn't work. How could Adobe have never programmed Flash with this addiction?

    In any case, it appears that in the cycle of upgrading Ubuntu insisted on bring back the HDMI audio in game. Which had been set correctly - off - before the upgrade. I have HDMI to my monitor, but its speakers are analog.

  • Oracle RAC upgrade 11.2.0.3 to 11.2.0.4

    Oracle installed:

    The RAC

    Oracle Database 11 g Enterprise Edition Release 11.2.0.3.0 - 64 bit Production, 2 nodes

    PL/SQL Release 11.2.0.3.0 - Production

    "CORE 11.2.0.3.0 Production."

    AMT for Linux: Version 11.2.0.3.0 - Production

    NLSRTL Version 11.2.0.3.0 - Production

    Linux: Red Hat 64-bit

    Question: How to 11.2.0.3.0 to 11.2.0.4.0 Oracle?   I need to create a new home?


    I'd appreciate any help.


    Thank you.

    Since it's CARS, you must first put the grid Infrastructure. Install IM 11.2.0.4 in a new home and have to do the upgrade at the same time. Once the GI is upgraded, you can then install 11.2.0.4 RDBMS as a CAR-install software into a new home. Then call the new home DBUA to upgrade db.

    It sounds simple when its put in this way, but each update needs sufficient tests. I also have a minimum span of two weeks, between IM and database updates.

    Here is a link to your version upgrade Guide: http://docs.oracle.com/cd/E11882_01/server.112/e23633/toc.htm

    HTH,

    Brian

  • 10g R2 no RAC-database unique Instance to upgrade to 11g R2-2 node RAC database

    Hi all

    I have a requirement for my client.

    We have a three databases with replication between them activated by MV & MV connects to the 10 g R2 database.

    This must be upgraded to Database 11g R2 with a 2 RAC node.

    Please tell us how to get started and go.

    Also please notify once upgraded to database 11g R2, how replication can be achieved.

    Thank you

    Vijayaraghavan.K

    > We have to first upgrade to 11 g R2 Instance unique and then move on RAC - node 2

    You must install the grid Infrastructure for Cluster and RAC database software in the new cluster.

    Then upgrade the database as a single instance database.

    Then, convert it to the RAC

    See http://docs.oracle.com/cd/E11882_01/rac.112/e17264/install_rac.htm#TDPRC185 for the conversion of the measures.

    Hemant K Collette

  • Steps/reference to upgrade node 4 RAC 11.2.0.2 to 11.2.0.4

    Hello

    I could not find note oracle good to know the steps of upgrade 11.2.0.2 to 11.2.0.4 on cluster of CARS for the node 7.1 4 AIX. Could some one note/links summary, if has?

    Upgrade of the GRID:

    Normally, we use to install 11.2.0.4 binary GRID at separate and then upgrade as 11.2.0.2 to 11.2.0.4. How can we do this and how do we upgrade all clustered components that point to the old GRID home (11.2.0.2) to newly installed HOME (11.2.0.4)?

    RDBMS upgrade:

    We install the new HOME for 11.2.0.4 and then we run carupgrade to upgrade the DB to 11.2.0.4. But how to change the cluster resources of RDMBS than HOME (11.2.0.2) of old points to 11.2.0.4.

    Thanks a lot for your suggestions and referring to any web link would be useful for me to read in more detail in this.

    Kind regards

    Hi kumarsk,

    your proposed update of grid and RDBMS are exactly what will be out of place update. The docs on this are available on the website of oracle as usual:

    http://docs.Oracle.com/CD/E11882_01/install.112/e41961/procstop.htm#CWLIN417

    http://docs.Oracle.com/CD/E11882_01/server.112/e23633/upgrade.htm#UPGRD003

    1. check that your environment is ready with cluvfy with a command similar to this:

    ./runcluvfy.sh stage pre - crsinst-upgrade to cars - n-test1, test2-rac-rolling - src_crshome /oracle/11.2.0/grid-dest_crshome /oracle/11.2.0/grid4-dest_version 11.2.0.4.0 - verbose

    2. run the installer of the grid 11.2.0.4 software and chose the upgrade option. That will install the software in a new home. Your old cluster will remain online until the point was that you run the script rootupgrade.sh on the first node. The upgrade continues to roll, your db will remain available with sessions of failover to other nodes. Do this for all 4 nodes and finish the Setup program later.

    11.2.0.4 handles transfer all files required including travel of Auditors. Of course, you should always apply some controls yourself from the environment to ensure that everything is just right.

    3. start the 11.2.0.4 db Installer choose only the software and install it in the new House.

    4. turn to level the databases one by one with dbua new home installed. The dbua will take care of registration of the db with the new values for oracle_home, etc in the clusterware.

    If you want to do the db upgrade manually, you can pass parameters in the cluster registry by running "srvctl upgrade database d xxx o new_oracle_home". You run this command from the database new home.

    Concerning

    Thomas

  • 11g R2 RAC on RHEL and upgrade the kernel

    Hi all

    We Hare seen 11 g r2 rac on RHEL 5 with 3 cluster nodes.

    We intend to upgrade OS RHEL (Version of the kernel 2.6.18-348. 1.1.el5 to 2.6.18 - 371.9.1.el5) one node at a time.

    Do we need to upgrade Oracle asm library to the same version of the OS kernel (current: 2.6.18 - 348)?

    Can you please let me know the the pre/post-prerequisites that need to be transported, or all documents/links detailing the steps.

    Kind regards

    Kumar

    See the MOS notes below:

    How to manage the Oracle grid Infrastructure during upgrades of OS (Doc ID 1559762.1)

    Reissue of links Oracle Home FAQ (frequently asked Questions) (Doc ID 1467060.1)

    How to upgrade driver ASMLib kernel as part of the upgrade of the kernel? (Doc ID 1391807.1)

  • Upgrade Oracle Rac 11.2.0.2 to 11.2.0.4

    Hello team

    We are about to upgrade a 2 node rac 11.2.0.2 to 11.2.0.4 and I need information on the methods to do this ensure minimum downtime.

    Does anyone have a step by step?

    Thanks in advance.

    Hello

    That is right.

    Kind regards

    Vautrin - Oracle

  • Steps to upgrade spirit RAC Standby Database?

    Hi all

    I RAC environment with 8 knots (4 Linux) I have a knot like physical standby for this CAR.

    My oracle is 10.2.0.3, and I want to update 10.2.0.5

    Could you advice I do not like, when I googled I found that I must begin by Clusterware before then?

    Any recommendation?

    Thanks in advance.

    Hello

    If you follow the second link I posted, all necessary measures are there, even if you have some CARS, it is explained here. Take a look at chapter "B.2 upgrade Oracle database with a physical Standby Database in Place", and if you have any questions post here.

    ARM

  • 11 GR 2 DBUA complains when upgrade no DB RAC server with other DBs CARS

    Hello

    I have encountered a problem using DBUA 11.2.0.3 yesterday and would like to know how to resolve or work around this problem.

    I am on 64 bit RH5 a cluster 2 node RAC with 11.2.0.3 infrastructure.  On node 1 of the cluster, I have a NON- RAC 11.2.0.2.1 database named DB1.  Note, this DB has ever been of CARS.  There is no other DBs running against this 11.2.0.2.1 home.

    "When I try to run DBUA to node1 using a separate 11.2.0.3 new House (with no database yet) against the DB1 database, it gives the following error message.

    The following requirement checks failed:

    The "ora {sid}" user does not have a configuration user equivalence on the nodes '[ud1 UD2]. "  Please set user equivalence before continuing.

    PRKC-1044: Impossible to verify the configuration of remote command for node [node 1] using/usr/bin/ssh and/usr/bin/rsh shells.

    [node1]: Connection refused

    You must meet these conditions before continuing the upgrade.  Do you want to continue?  Yes or no (buttons)

    ---

    Given that I am not upgrading a database RAC or using an Oracle Home in use by all instances of RAC, in my view, that this is a bug in DBUA.

    Anyone encountered this?  How do you get around that?

    ---

    I have an SR is open.  The database serves as an EBS 11i environment and I am following docid 881505.1.  I asked support if do manual upgrade DB (without DBUA) is a workaround.  I await their response.

    Thank you, Dean


    Support of Oracle says to do the upgrade manually.  DBUA confuses because we installed on the node to upgrade network infrastructure, but the DB to be upgraded on the node is not cars.  I was able to complete the upgrade manually.  I'll ask management to create a bug for this.

    Dean

  • upgrade to oracle 9i to Oracle 11 g RAC RAC

    Hi all

    version of DB: oracle 9.2.0.5 RAC
    OS: Linux

    My project team and the customer plans to upgrade oracle 9i RAC to Oracle 11g RAC.
    Please suggest any MOS notes all the links.

    Thank you
    Visu.

    Published by: visu996253 on 8 April 2013 05:30

    version of DB: oracle 9.2.0.5 RAC
    OS: Linux

    My project team and the customer plans to upgrade oracle 9i RAC to Oracle 11g RAC.
    Please suggest any MOS notes all the links.

    Note to master for the upgrades of Oracle database and migration [ID 1152016.1]
    http://www.Oracle.com/technetwork/products/upgrade/index-088044.html
    785351.1 - companion of oracle 11g upgrade
    http://download.Oracle.com/docs/CD/E11882_01/server.112/e17222.PDF - Guide to upgrade the database® Oracle 11 g Release 2 (11.2) E10819-02

    Published by: KR10822864 on 8 April 2013 05:52
    Add more information.

  • RAC upgrade databases on grid of 11.1 to 11.2 11.2

    Hello

    I have a database of 2 node rac with oracle grid infrastructure and ASM = 11.2.0.2 version.
    The database is an earlier version 11.1.0.7. (linux Redhat 5.4 intel)
    We installed this way because applications running on this database was not
    certified in these times with 11.2.
    Now, we want to upgrade the database and let the version of the game IM and RAC.

    Is it possible to upgrade the Rac database without stopping the services?
    I read documents on ASM rolling upgrade (not always possible, but only in)
    (in some cases).
    Is there something like an upgrade of 11.1 to 11.2 database?

    Thank you.

    11.1 to 11.2 is an upgrade via DBUA, this is not a patch. See pl

    Changes important related to Patch Oracle database sets starting with 11.2.0.2 [ID 1189783.1]
    Quick reference to group of patches Patch [753736.1 ID] numbers

    HTH
    Srini

  • Document on 11 GR 1 material upgrade to GR 11, 2 RAC RAC

    Friends,

    I have install RAC 11 GR 1 material in vmware with the common house for ASM and data base. I intend to upgrade the installer to GR 11, 2 CARS.

    If you have the document on the upgrade process, please share with me. I tried Google, but did not get any correct document.

    Thanks in advance,
    Florine

    http://docs.Oracle.com/CD/E11882_01/install.112/e10812/procstop.htm

    Update 11.2.0.1 to 11.2.0.3

  • RAC 11.2.0.1 to 11.2.0.2 to the new Cluster upgrade with any number of nodes

    I'd like suggestions on upgrading a RAC Oracle 11.2.0.1 in Linux x86_64 RHEL 5.3 with 6 cases of a new cluster with 4 linux x86_64, RHEL 5.5 and upgrading to 11.2.0.2 nodes group of hotfixes.
    New SAN storage will be used. Both systems use ASM.

    Current system... New system
    11.2.0.1... 11.2.0.2
    6 instances... 4 instances
    Linux RHEL 5.3 x86_64... Linux RHEL 5.5 x86_64

    Method:
    (1) run controls and upgrades pre scripts
    (2) save the current database with RMAN and put at the disposal of the backup files to the new host
    (3) install grid Infrastcure on the new cluster 11.2.0.2
    (4) set up ASM on the new Cluster
    (5) install the Oracle RAC 11.2.0.2 on the new cluster database and create the database with the same name of the original database.
    (6) startup nomount using a parameter file and the source of DBID edited without the cars parameters (cluster_database = false, etc...)
    7-restore Rman backup controlfile to step 1
    (8) restore / restore incomplete database
    (9) OPEN RESETLOGS UPGRADE
    10) proceed to updates manual steps (script catupgrd.sql, etc.)
    (11) remove the cancellations additional tablespaces e redo log son
    (12) CONVERTED to THE CARS USING DBCA or RCONFIG

    This method is correct? Do you offer other methods or corrections?

    I also want to know if there is another thing to delete, besides redo threads and undo tablespaces, due to the smaller number of cases.

    I also considered using RMAN DUPLICATE to the same version (11.2.0.1) and then MOVE to 11.2.0.2, but DOUBLE would change the DBID.

    Thank you

    Hello

    your efforts seem valid, but its awfully difficult.

    Why not simply create a watch on 11.2.0.1 and then update.
    If you are using "duplicate for standby" should not change the dbid.

    Then, you can just 'turn on' the night before and continue the upgrade.

    Concerning
    Sebastian

  • Upgradation OS for RAC

    Hi guys... How is everything...

    I have the node CCR on Windows2003R2 Itaniam. I need to know is there any procedure for upgradation OS for RAC... I mean I'm going to the bottom of the Cluster & when the OS database places a gradation.

    Hello

    you need not stop the entire base.
    Upgrades of the OS can be in upgradeable mode (so one node after another), you must stop the instance and the clusterstack on the node, you perform the upgrade.

    Policies to support Oracle to launch of the CARS with different versions of an operating system within a period of 24 hours.
    In other words: after completing the upgrade of the first node you have 24 hours to what your other nodes must be at the same level of the new OS.

    Concerning

    Sebastian

Maybe you are looking for

  • Start of message on my Satellite A50 101

    My Satellite laptop displays a message after the start. It is related to "ZeroCfgSvc MFC"I saw something on the device wireless in some forums. But my wireless connection works well... And it's embarrassing to see that whenever I use the computer.Do

  • Installed the library at Ext Drive, iTunes wants to re-download all previous purchases

    I moved the library iTunes & iPhoto on an external drive, as described in the guide online from Apple. Everything seems to work properly. Then I upgraded my OS to El Capitan, and I try to sync our iPhones and iPad for the first time. In iTunes, is to

  • What is this problem with Verizon abandoned Microsoft Premium

    Verizon reported that they no longer Microsoft Premium with their dsl.  What that means for me?  electronic mail, Favorites, contacts. etc /

  • Windows 7 quick launch buttons - Reg

    Hello I use HP Pavilion laptop dv5-1102TU. I have recently upgraded to Windows 7 Ultimate from Windows Vista Home Basic by results of Windows 7 upgrade advisor. After the installation of Windows 7 quick launch buttons work for me, but light does not

  • EqualLogic SAN and Jumbo frames

    I can't get the jumbo frames to work on our Equallogic PS6000.  We have the following configuration PS6000-> Dell PowerConnect 6248 (bunk)-> Windows 2008 Server. The network cards on the SAN has been set at 9000 MTU, ports on switches are set to MTU