Hetrogenous RAC - is it possible ?

Is it possible to compose a RAC Cluster using nodes runing different OS's ( I suspect the answer is no ) but I want to tease out a couple scenarios
I recall challenging Oracle on their Grid concept (I suppose really now replaced with their cloud concept ) which was lauded as being able to add in any spare servers to consolidate processing power. I then asked the question could you therefore combine a Windows Server with unix server in regards to the processor grid. In pure grid theory the answer should be yes but when pressed Oracle they admitted no.
Q1. So I am guessing the same is true of RAC - I suspect you could not have one node as Windows and another node as unix ( the difference in the file path specification alone would surely scupper it ).
Q2. What about a unix node and a linux node ?
Q3. What about binary compatible nodes eg one node being RHEL and the other being OEL ?
Q4. Finally what about 2 nodes that were the same OS but had a slight different version of ORACLE_HOME eg 11gR1 and 11gR2 ?
any thoughts appreciated,
Jim

Jimbo wrote:
I recall challenging Oracle on their Grid concept (I suppose really now replaced with their cloud concept ) which was lauded as being able to add in any spare servers to consolidate processing power. I then asked the question could you therefore combine a Windows Server with unix server in regards to the processor grid. In pure grid theory the answer should be yes but when pressed Oracle they admitted no.RAC is not Cloud. Despite what the marketing suits may spin.
Q3. What about binary compatible nodes eg one node being RHEL and the other being OEL ? Yes. In fact did this recently on a test cluster. Not a good idea though (obviously) as there can be some kernel driver issues (e.g. ADVM of 11.2.0.2 runs on 5.4 kernel and not on the 5.7 kernel). Also makes o/s management more complex. I'm a big believer that each and every server node in a Grid/RAC needs to be identical at o/s level. Same kernel version, same drivers, same rpm's, same +/etc+ configs (such as hosts, multipath, udev) and so on.
Q4. Finally what about 2 nodes that were the same OS but had a slight different version of ORACLE_HOME eg 11gR1 and 11gR2 ?Well, you can have multiple Oracle Homes on the same server. So you should be able to have a 10 node Grid, policy managed RAC, server pool 1 running a 10gr2 RAC database as 5 instances, and server pool 2 running a 11gr2 RAC database as 5 instances. Offhand I do not see why this would not be possible.
Some years ago (due to circumstances) I had another Oracle Home on node 1 of a 4 node RAC, running a standalone Oracle db instance on the RAC sever node that was running RAC db instance 1. (since has moved that to its own dedicated server). So I do not see why this 2nd Oracle Home could not run another RAC db version...
any thoughts appreciated,Well, the obvious thought is why do you want to throw heterogeneous stuff together into an Oracle RAC? To achieve what need, and address which requirement, exactly?

Similar Messages

  • How many maximum nodes possible in RAC 10g

    Hi,
    This is for my knowledge only.
    Any idea that how many RAC nodes is possible in 10g maximum.
    Thanks and regards,
    Chitrasen

    Hi Chitrasen :
    From MetaLink:
    "How many nodes are supported in a RAC Database?:
    With 10g Release 2, we support 100 nodes in a cluster using Oracle Clusterware, and 100 instances in a RAC database. Currently DBCA has a bug where it will not go beyond 63 instances. There is also a documentation bug for the max-instances parameter. With 10g Release 1 the Maximum is 63. In 9i it is platform specific due to the different clusterware support by vendors. See the platform specific FAQ for 9i."
    So, for 10g R2, looks like 63... lots more than most sites will ever use, I suspect!
    HTH.

  • It is possible to apply redo while the standby is open in read only mode ?

    Hi,
    I am using Oracle 11g R2 (11.2.0.1.0)
    Step 1: Two node RAC is Configured.
    Step 2: One node Data Guard RAC is configured.( i,e Standby is one node RAC).
    Primary:
    SQL> select open_mode,database_role,PROTECTION_MODE,PROTECTION_LEVEL from v$database;
    OPEN_MODE DATABASE_ROLE PROTECTION_MODE PROTECTION_LEVEL
    READ WRITE PRIMARY MAXIMUM AVAILABILITY MAXIMUM AVAILABILITY
    Standby:
    SQL> select open_mode,database_role,PROTECTION_MODE,PROTECTION_LEVEL from v$database;
    OPEN_MODE DATABASE_ROLE PROTECTION_MODE PROTECTION_LEVEL
    READ ONLY WITH APPLY PHYSICAL STANDBY MAXIMUM AVAILABILITY MAXIMUM AVAILABILITY
    Question:
    I have not purchased active dataguard license.
    In my case In Data Guard RAC it is possible to apply redo while the database is open in read only mode?
    Thanks
    Solaiman

    876149 wrote:
    SQL> select open_mode,database_role,PROTECTION_MODE,PROTECTION_LEVEL from v$database;
    OPEN_MODE DATABASE_ROLE PROTECTION_MODE PROTECTION_LEVEL
    READ ONLY WITH APPLY PHYSICAL STANDBY MAXIMUM AVAILABILITY MAXIMUM AVAILABILITY
    "READ ONLY WITH APPLY" in V$DATABASE.OPEN_MODE means Active Data Guard is enabled.
    Oracle code cannot check if you have the right license: it is up to you to know which license you have for your database environments.

  • Oracle 9i Upgrade to 10g - Single Database instances to RAC

    Hi,
    We are considering migrating from single SAP database architecture into
    an Oracle 10g RAC architecture. The question i have is:
    Are multiple PHYSICAL databases supported in one Oracle 10g RAC environment? For example, Is it possible to install and operate multiple databases in one single ORACLE 10g Real Application Cluster such as one SAP R/3 database, one SAP EBP and one
    SAP BW database co-exist in a single Oracle RAC environment?
    Regards,
    Chengappa Ballachanda

    Hi Sairam,
    Due to the size of our Production environments we have decided not to pursue the MCOD option.
    Eg: Take a QA environment which consists of SRM and BW and we have multiple nodes with both installed on each node. Now if I were to go the MCOD option then these instances can connect to one large Oracle File System using different schema's. 
    What I'm trying to establish is in that one RAC, is it possible to have the multiple database instances on each node (SRM, BW etc) connect to multiple Oracle File Systems ie SRM connect to its own OFS and BW to its own instead of the one large shared OFS?
    Hope this makes sense! Easier if I could upload a diagram!
    Regards,
    Chengappa

  • Oracle 9i Standalone to 10g RAC ??

    We have Oracle 9i on Solaris 9 with about 2 TB of data/index on SAN. We want to migrate this to 2 new boxes running Solaris 10 and Oracle 10g RAC -- storage SAN. What should be the best approach considering:
    1. Fastest copying method for data/index from existing system (Solaris 9/Oracle 9i -> Solaris 10/Oracle 10g -- all on separate machines).
    2. No return to 9i, once it 10g, stay Stanalone and/or RAC.
    3. If 10g Standalone to RAC creates problem, stay on 10g Standalone and try later.
    4. If ASM creates any issue, stay non-ASM/RAW files and try later.
    Thanks.

    Hi Sairam,
    Due to the size of our Production environments we have decided not to pursue the MCOD option.
    Eg: Take a QA environment which consists of SRM and BW and we have multiple nodes with both installed on each node. Now if I were to go the MCOD option then these instances can connect to one large Oracle File System using different schema's. 
    What I'm trying to establish is in that one RAC, is it possible to have the multiple database instances on each node (SRM, BW etc) connect to multiple Oracle File Systems ie SRM connect to its own OFS and BW to its own instead of the one large shared OFS?
    Hope this makes sense! Easier if I could upload a diagram!
    Regards,
    Chengappa

  • 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

  • Using native APEX on 11gR2 RAC

    Hi all,
    I'm not sure if this is the right forum for this question, please forgive me if it isn't.
    When using native APEX on 11gR2 RAC, is it possible to balance connections between instances ? If yes, how ?
    I've successfully installed 11gR2 RAC, and configured APEX, and I'm able to use it, but it runs on one instance only.
    Thanks in advance
    Antoine

    Hi Antoine,
    How are you connecting to the database, via the Apache HTTP Server mod/plsql connector or are you using the embedded PL/SQL gateway (EPG)?? In the first case connecting to the SCAN listener should balance the connections between the instances if they are both registered for the service you connect to. I'm not sure how the EPG works under RAC. In general, I'd prefer the application to only connect to one of the RAC instances by default and only transfer across due to failover.
    Rob..

  • RAC Help

    I have two RHEL 4.0 servers with RAID Enabled with 2 NICs and 1 GB RAM. I want to install oracle 10g on these machines with 2 node RAC. is it possible? what i referred from ORACLE Press book is a three node demo. How to setup RAC on two nodes...I do not have a third machine. Kindly advise. regards ....

    You don't need a thrid machine to setup and RAC.
    Two machines will be enought to setup the RAC setup. Follow the instruction in the following link.
    http://www.oracle.com/technology/pub/articles/hunter_rac10gr2.html
    - Leo

  • Is it better to use Invoke nodes or property nodes to set/get control values?

    I have a series of VI's that run in parallel, each to manage different functions- pumping, sensing, a fluid flow model, an experiment generator/runner.
    These need to exchange data, which I am currently doing using invoke nodes (that are all in subVIs), using methods "Set control value" and "Get control value". I find that every now and then (perhaps 1% of the time) the data isn't exchanged correctly and therefore the system doesn't work. I can imagine how "set" could go wrong if they happen simultaneously, and can devise ways of preventing this. However, the "Get" method suffers from the same problem. This is a major problem, because I want to leave it running for several hours.
    I could in
    principle achieve the same thing using property nodes and find myself wondering if this might be more reliable. But I don't want to change over only to find it makes no difference!
    Can anyone advise?

    You can use some kind of syncronization such as queues, occurances, or notifiers but I think the easiest way would be to create and action engine. This was only one action can execute at a time avoiding a "race condition". Possibly a write action and a read action might help. You also can add queues or notifiers into this concept. hope this helps.
    BJD1613
    Lead Test Tools Development Engineer
    Philips Respironics
    Certified LV Architect / Instructor

  • How to restrict login for multiple users having same Role

    Our Web Application is deployed on Tomcat 5.5
    The requirement is ?
    There are roles in application like "operator", "admin"?
    There are multiple users created for each of the above role.
    When one user of "operator" role is logged in, then
    It should not allow to login for another user of "operator" role.
    Also, if user did not log out & application gets close, then
    It should not allow to login for another user of "operator" role.
    Also, it should not allow to login for multiple requests of same user
    (using another browser instance...)
    Is it possible using session object?
    But, using session object, it will create separate objects for different users,
    So here I will not be able to restrict session object creation rolewise.
    Also, how to retrieve these multiple session objects created for different users on server?
    If anyone is having the solution please reply as soon as possible,
    Thank you.

    To tell you the truth, this is a stupid requirement. It must be an extremely fragile application.
    In any case, you will have to write your stuff for that. Probably a filter that on login, logout, and session expiration checks, makes, or removes entries in a DB (using a synchronized resource to prevent race conditions) or possibly even simply in an application context object.

  • A tough performance problem after upgrade to 10g

    A stored procedure is running fine (returns in 2 min) against SQLPLUS. But when it is called from the application interface, it will take 40 min to return. I have traced the session and found there is one query takes most of the time waiting on 'db file sequential read', 'gc cr grant 2-way', 'gc cr block 2-way', etc. Is there something needs to be tuned from the database point of view. I cannot explain why it's running perfect in SQLPLUS.

    Are the query plans the same when the procedure is called from SQL*Plus and when it is called from the application?
    What proportion of the wait time is attribued to the "gc cr grant 2-way" and "gc cr block 2-way" events? Those are events caused by cache fusion in Oracle RAC-- is it possible that the application is running on a different node than your SQL*Plus session? If those events are a significant fraction of the overall wait time, that implies that some other instance has the block pinned. Do you have a handful of hot blocks that have to get sent between different instances constantly?
    Justin

  • Oracle 11g r2 report server on vmware esxi 5.1

    dear gurus
    assist me we are facing issue with oracle 11g r2 report server on vmware esxi 5.1
    see all of sudden report function, however we restart server then it work fine,
    where it is on windows 2008 r2 sp1
    our vendor say not certify by oracle this product, we have no rac just standalone env
    1db 1 form 1 report server  all on VMs guest working
    we could not find os vm oracle logs root causes
    kindly assist or advise

    user6029797 wrote:
    Now - I wanted to create 2 such VM Guest Linux O/S instances and make those 2 Linux/Oracle into Oracle 11g R2 RAC
    Is this possible to do - so I can learn about Oracle 11g R2 RAC operation, tuning, setup and maintenance?Yes, possible to setup.
    No, you are unlikely to learn anything of value ito tuning and maintenance as the environment does not run on actual cluster h/w that are used in real world RAC clusters. For example, huge pages, jumbo frames - pretty meaningless in a VM for tuning Oracle RAC. Bonding of Interconnect Interfaces and pulling the plug - pretty much useless. What about running multiple I/O paths to the cluster's storage server? Using RDS for Interconnect instead of UDP? Etc. Etc.
    RAC is only as performant, scalable and robust as the h/w components it is build on. I cannot see any value in virtualising this and expecting to gain real insight to RAC performance management and tuning. Best one can expect is familiarisation with installation prerequisites, the installation steps and setting up the database. I question if any more value than that can be obtained from a single PC virtualised RAC environment.

  • Replication argh, which is better for this setup

    Hi,
    I'm a bit lost here would appreciate some input.
    Assume it will be running 11g R2 Redhat or Oracle Ent Linux
    This is the config
    RAC1:
    RAC
    2 x Standard Edition, some sort of replication (small, couple of tables). both servers are identical just there for redundancy.
    RAC2:
    RAC
    2 x Standard Edition, some sort of replication (small, couple of tables). both servers are identical just there for redundancy.
    I don't know much about data guard but it sounds like its the best way to go for Oracle?
    I would like to only have 1 x SE at each RAC, is this possible?
    Any guidance would be great, thank you.

    Hi,
    Dataguard is not Avl. for SE.
    Check this Ref Doc:Getting Started with Data Guard
    2.3.2 Oracle Software Requirements
    The following list describes Oracle software requirements for using Data Guard:
    Oracle Data Guard is available only as a feature of Oracle Database Enterprise Edition. It is not available with Oracle Database Standard Edition.
    HTH

  • Diff between Oracle 9i Standard Editionand Oracle 9i Enterprise Edition

    what is the diff between Oracle 9i Standard Edition     and Oracle 9i Enterprise Edition

    Check http://www.oracle.com/database/product_editions.html
    This is about 10g but mostly should be the same as in 9i with at least two exceptions known for me RAC is not possible for 9i SE and 9i had not XE at all.
    Gints Plivna
    http://www.gplivna.eu

  • Wrong graphic at HTML-report

    Hello!
    In a first sequence I export a graphic to a HTML-report, which I close afterwards. In the second sequence I export an other graphic (generated by the same SubVI, so same size/looking (polar plot or XY-graph), but different data and other picture-variable at each sequence) to another HTML-report , but sometimes (not always) LabView 7express adds the graphic from the first report to the second one. The second picture-file (jpg) was not created then, but the link to the picture in the second HTML-report referes "technical correct" to the wrong (first) picture.
    If I export the reports twice (without changing anything), the problem disappears in most cases.
    Do you have any idea how I can solve this problem?
    Thank you!
    martin

    This sounds like a type of race condition or possibly a problem with multiple pointers to the same memory location.  Without seeing some code, it will be very hard to troubleshoot.  Could you post an example that show off this problem?
    Thanks,
    Bob Young
    Bob Young - Test Engineer - Lapsed Certified LabVIEW Developer
    DISTek Integration, Inc. - NI Alliance Member
    mailto:[email protected]

Maybe you are looking for

  • Upgrade to FCP 5.1.4 and OSX 10.4.9 stops DV

    Hi all, Have upgraded to from OSX 10.3.9 to OSX 10.4 to allow me to upgrade FCP from 4.5 to 5.1 in order to add HDV support. First use after upgrading and FCP is no longer recognising DV camera through Firewire. Subsequently upgraded to OSX 10.4.9 an

  • Document should not splitt to tax line items

    Hi, Document should not splitt for tax line items,currently it is happining how contol this. Thanks and Reg

  • Generate sequence nuumber for line items.

    Hi, Am using jdeveloper 11.1.1.6.0 I have generated a form using master detail. po_header as master and po_line as child in po_line i need to generate the sequence number. The problem am facing is If i generate using create sequence in my DB. Its wor

  • Trying to delete a cost centre against a position

    Hello fellow SAP users, We've run into a issue when trying to delete a cost centre record against a position, the scenario is as follows: Cost centre was created on the Parent Org Unit on the 01.01.1900 and delimited 01.06.2010. This created an infot

  • Finger print

     my finger print scanner does not work on the lock screen but when i log in using the password it will then work.  i'm trying to open digital persona personal there is user name and password i'm tring to type same password not accepted . kindly advic