Questions on 10g AS Clustering

I worked on non-cluster AS environment but not on cluster environment and need your help/advice on this. Suppose I have 5 machines:
One machine will have Metadata/Database.
Two machines will be part of WebCache/SSO Cluster.
Two machines will be part of OC4J/OID Cluster.
Where & how can I know details on clustering, e.g. how to implement this?
I have UNIX systems.
Thanks.

These articles will get you started for oracle app server clustering.
First read the first article and figure out which archtecture fits your envirnoment. http://www.oracle.com/technology/pub/notes/technote_sowa.html
http://download.oracle.com/docs/cd/B14099_07/core.1012/b13998/toc.htm

Similar Messages

  • Oracle 10g DB clustering question

    Hi All
    If I am going to cluster my db, what method/type of software should I use.

    Hi,
    The most suitable s/w changes by the thing which your DB regards as important.
    System requirements (ex SLA) , cost ...
    Oracle RAC is Acitve-Active cluster solution.
    All of node is active and can execute a sql.
    However, the license cost becomes very higher.
    I may be enough in Active-Stanby if a downtime is permitted at the time of instance failure.
    PowerHA(AIX)
    MC/ServiceGuard(HP-UX)
    SunCluster(Solaris)
    MSCS(Windows)
    ...

  • SAP Adapter Best Practice Question for Deployment to Clustered Environment

    I have a best practices question on the iway Adapters around deployment into a clustered environment.
    According to the documentation, you are supposed to run the installer on both nodes in the cluster but configure on just the first node. See below:
    Install Oracle Application Adapters 11g Release 1 (11.1.1.3.0) on both machines.
    Configure a J2CA configuration as a database repository on the first machine.
    Perform the required changes to the ra.xml and weblogic-ra.xml files before deployment.
    This makes sense to me because once you deploy the adapter rar in the next step it the appropriate rar will get staged and deployed on both nodes in the cluster.
    What is the best practice for the 3rdParty adapter directory on the second node? The installer lays it down with the adapter rar and all. Since we only configure the adapter on node 1, the directory on node 2 will remain with the default installation files/values not the configured ones. Is it best practice to copy node 1's 3rdParty directory to node 2 once configured? If we leave node 2 with the default files/values, I suspect this will lead to confusion to someone later on who is troubleshooting because it will appear it was never configured correctly.
    What do folks typically do in this situation? Obviously everything works to leave it as is, but it seems strange to have the two nodes differ.

    What is the version of operating system. If you are any OS version lower than Windows 2012 then you need to add one more voter for quorum.
    Balmukund Lakhani
    Please mark solved if I've answered your question, vote for it as helpful to help other users find a solution quicker
    This posting is provided "AS IS" with no warranties, and confers no rights.
    My Blog |
    Team Blog | @Twitter
    | Facebook
    Author: SQL Server 2012 AlwaysOn -
    Paperback, Kindle

  • Good questions on 10g

    i need some information on what types of questions come in 10g which a 10g person should be knowing.
    Any web-site or documentation will be helpful.
    I hope, my question is clear. Please help, in solving the doubt.
    regards.

    Hi,
    As Frank just told you.
    Here you have the link to get every doc. about AS
    http://www.oracle.com/technology/documentation/appserver.html
    I hope U have enough on this information otherwise let me know.
    Cheers,
    Hamdy

  • Frmwebutil.jar Question w/ 10g (9.0.4)

    All over the Oracle documentation w/ Dev Suite and Webutil configurations it refers to frmwebutil.jar and sometimes f90webutil.jar.
    My question is, are we to use both files in various places w/ 10g (version 9.0.4) or use one or the other exclusively?
    If we are to use f90webutil.jar then I assume every reference directing us to frmwebutil.jar, we replace w/ f90webutil.jar, including signing the jar files adding to registry, .cfg files, etc...correct?
    thank you.

    Hello,
    using one of them is enough,Oracle Developer Suite 10g Release 2 (10.1.2) uses frmwebutil.jar. and the earlier release uses dev suite f90webutil.jar
    I think in your case you can use frmwebutil.jar alone.
    Regards
    Mohan

  • Question: 9i- 10g

    Hi all,
    What does anyone know or have an opinion regarding 10g's ability with Virtual Private Databases/Row level security and Unix based servers providing client-server forms and reports services to browser based windows and unix clients?
    My organisation has been stuck in Forms 6 land because of heavy limitations of Forms 9's ability to handle row level security in this situation and I was curious to know if anyone has upgraded to 10g and resolved any issues in that regards.
    Thanks all.

    This is another article:
    TECHNOLOGY: Security
    Keeping Information Private with VPD
    By Arup Nanda
    Oracle's row-level security gives users their own virtual private databases.
    Ensuring appropriate information privacy is a pressing concern for many businesses today, given privacy legislation such as the United States' HIPAA (Health Insurance Portability and Accountability Act), Gramm-Leach-Bliley Act, Sarbanes-Oxley Act, and the EU's Safe Harbour Law. Other privacy mandates, such as Visa's Cardholder Information Security Program (CISP), also require businesses to ensure that access to information is tightly controlled.
    Oracle has always included the ability to grant (or deny) users access to database objects, but these privileges are defined at the object level—for an entire table, not for specific rows in that table. Although that approach is sufficient for many applications, any application touching on financial, health, or other kinds of personal information usually requires more-discrete controls over access and authorization.
    Oracle's row-level security (RLS) feature, introduced in Oracle8i, provides fine-grained access control—fine-grained means at the individual row level. Rather than opening up an entire table to any individual user who has any privileges on the table, row-level security restricts access to specific rows in a table. The result is that any individual user sees a completely different set of data—only the data that person is authorized to see—so the overall capabilities are sometimes referred to as Oracle's virtual private database, or VPD, feature.
    Using Oracle's VPD capabilities not only ensures that companies can build secure databases to adhere to privacy policies but also provides a more manageable approach to application development, because although the VPD-based policies restrict access to the database tables, they can be easily changed when necessary, without requiring modifications to application code.
    For example, say a bank's account managers (AMs) provide personal customer support to high-net-worth account holders. AMs use a custom banking application to help them check their customers' balances, deposit or withdraw funds, and decide on loan requirements, for example. At one time, the bank's policy was to allow all AMs to access all account holder information, but that policy was recently changed. Now, AMs are assigned to a particular set of customers, and they need to be able to access information pertaining only to those customers. The policy change needs to be reflected in the application, which currently shows all customer information to each AM, not just information on the customers to whom each particular AM is assigned.
    To make the application comply with the new privacy policy, the bank has three choices:
    Modify the application code to include a predicate (a WHERE clause) for all SQL statements. This option doesn't ensure privacy policy enforcement outside the application, however, and if there are other changes in the future, the code will once again have to be modified, so this is not a good approach in the long term.
    Leave the application intact, creating views with the necessary predicates and creating synonyms with the same name as the table names for these views. This option is better from the perspective of application changes and security, but it can be difficult to administer, because of the potentially large number of views to track and manage.
    Create a VPD for each of the AMs by creating policy functions that generate dynamic predicates, which can then be applied across all objects, regardless of how they are accessed, by setting up policies with the row-level-security built-in package (DBMS_RLS).
    This last option offers the best security without administrative overhead and ensures complete privacy of information—all the account managers see a different view of the table, according to their own credentials.
    This article shows you how to set up a VPD security model. It goes through the process by using the foregoing bank scenario to create policy functions, define policies, and then test results. (Note that the tables are not completely defined, to keep the example simple.)
    Basic Setup for the Example Application
    Briefly, here are the basic assumptions about the example bank application:
    A BANK schema owns two key tables that make up the application—a CUSTOMERS table:
    Name Null? Type
    CUST_ID NOT NULL NUMBER
    CUST_NAME NOT NULL VARCHAR2(20)
    and an ACCOUNTS table:
    Name Null? Type
    ACCT_NO NOT NULL NUMBER
    CUST_ID NOT NULL NUMBER
    BALANCE NUMBER(15,2)
    Listing 1 comprises the SQL script for creating and populating these two basic example tables.
    User SECMAN (security manager) owns an ACCESS_POLICY table that identifies the AMs and their respective customer accounts:
    Name Null? Type
    AM_NAME NOT NULL VARCHAR2(20)
    CUST_ID NOT NULL NUMBER
    ACCESS_TYPE NOT NULL CHAR(1)
    The AM_NAME column stores the user ID of the account manager; CUST_ID identifies the customer; and ACCESS_TYPE defines the specific access entitlement—S (SELECT), I (INSERT), D (DELETE), or U (UPDATE). Some example records from the ACCESS_POLICY table follow:
    AM_NAME CUST_ID ACCESS_TYPE
    SCOTT 123 S
    SCOTT 123 I
    SCOTT 123 D
    SCOTT 123 U
    SCOTT 456 S
    SCOTT 789 S
    LARA 456 I
    LARA 456 D
    LARA 456 U
    LARA 456 S
    As you can see, SCOTT, the AM for customer 123, has all privileges—S, I, D, and U—as does LARA for customer 456. Also, because SCOTT can confirm account balances for one customer of LARA, he has S privileges on customer 456.
    Step 1. Create a Policy Function
    The bank's access rules (contained in the ACCESS_POLICY table) must be applied somehow, dynamically, whenever an AM tries to look into customer account information. The first step is to create a policy function that returns the appropriate predicate to be applied to the table. As with the ACCESS_POLICY table, the policy functions are created and controlled by user SECMAN, for security purposes. It's best to keep privacy rules separate from the tables to which they apply.
    Let's examine the policy function, get_sel_cust_id—shown in Listing 2— in detail:
    Accepts exactly two parameters: the schema name (p_schema in varchar2) and the table name (p_table in varchar2)
    Returns one value only, a string—return varchar2 as l_retstr varchar2(2000);—comprising the predicate that will be appended to every query on the table
    Uses a cursor routine—for cust_ rec in—to get the list of values, because each user may have several cust_ids listed in the table
    Returns a constructed string—l_retstr—to be used as a predicate whenever any user attempts to access the underlying table
    The function returns the predicate where cust_id in with the appended list of customer accounts (cust_id), separated by commas, that the user (am_name = USER) is allowed to see.
    Note that before entering the loop that builds the list of cust_ids for the user, the code in Listing 2 checks to see if the user is the table owner BANK, in which case the function returns a NULL predicate, as follows:
    if (p_schema = user) then
    1_retstr := null;
    After building this function, you want to make sure it returns the appropriate predicate, by testing some sample data. Connect to the database as SECMAN, and insert some records into the ACCESS_POLICY table, giving SECMAN read privileges on a few sample accounts, as follows:
    insert into access_policy values ('SECMAN',123,'S');
    insert into access_policy values ('SECMAN',456,'S');
    insert into access_policy values ('SECMAN',789,'S');
    Now execute the function:
    select get_sel_cust_id
    ('BANK','CUSTOMERS') from dual;
    The function returns a string that will be applied as a predicate, as shown in the following sample output:
    GET_SEL_CUST_ID('BANK','CUSTOMERS')
    CUST_ID IN (123,456,789)
    You need to create similar functions for the other types of access. For simplicity's sake, create a single function for all the other access types—UPDATE, DELETE, INSERT—as shown in Listing 3. However, note that for a real-world application, each type of access should have its own individual function defined, to ensure appropriate privacy.
    The policy function in Listing 3 is nearly identical to the policy function in Listing 2, except that the predicate is further qualified by use of the information from the ACCESS_CONTROL table:
    and access_type in ('I', 'U', 'D')
    Creating a policy function is just the first step. You now need to ensure that the function will be used, by defining the policy that should control its use in your system.
    Step 2. Define a Policy
    Policies are defined with the DBMS_RLS package, which is Oracle-supplied. Be aware that the policies themselves are not database objects owned by any user (schema); they are logical constructs. Any user who has the execute privilege on the DBMS_RLS package can modify or drop a policy created by another user. Privileges to DBMS_RLS should be judiciously controlled and granted with caution.
    In the following example, user SECMAN is granted execute privileges (by SYS) on the DBMS_RLS package:
    grant execute on dbms_rls to secman;
    Listing 4 creates a policy named CUST_SEL_POLICY on the table CUSTOMERS of schema BANK. This policy applies the predicate returned by the function GET_SEL_CUST_ID (which is shown in Listing 2) owned by schema SECMAN to all SELECT statements on the table.
    Similarly, you place another policy on the table for other access types, as shown in Listing 5. This policy applies to inserts, updates, and deletes in the CUSTOMERS table.
    It is almost identical to the SELECT policy, except that this policy includes a check that ensures that the policy will remain compliant even after an update:
    update_check => TRUE
    Data cannot be added to the table unless it adheres to the policy.
    Step 3. Test the Setup
    Now that the building blocks are in place, let's see how they work. Connecting as user BANK and issuing a simple select * from customers; query displays the following:
    CUST_ID CUST_NAME
    123 Jay Kulkarni
    456 Wim Patel
    These two records are the full contents of the CUSTOMERS table, and both records are shown because BANK owns the table, so the predicate clause is NULL—that is, no predicate is applied. However, when user LARA makes the same query, she sees the following:
    select * from customers;
    CUST_ID CUST_NAME
    456 WIM PATEL
    LARA sees only CUST_ID 456, not 123, because that is the row she is authorized to see, as determined by the ACCESS_ POLICY table. Note that the query has no WHERE clause but that the selection from the table is automatically filtered to show only the authorized rows.
    If user SCOTT makes the same query, his results are different from the results for LARA: select * from customers;
    CUST_ID CUST_NAME
    123 Jay Kulkarni
    456 Wim Patel
    User SCOTT sees both rows, because he is authorized to do so, as shown in the ACCESS_POLICY table. When user LARA issues the query, the policy function get_sel_cust_id returns the predicate where cust_id in (456). Lara's original query select * from customers is rewritten as
    select * from
    (select * from customers)
    where cust_id in (456)
    The predicate is automatically appended to the user's original query. The same thing happens when the user updates the table:
    SQL> update bank.customers
    2 set cust_name = 'PAT TERRY';
    1 row updated.
    Note that in this example, only one row is updated, even though there are actually two rows in the underlying table. The policy (CUST_IUD_POLICY) appends the predicate where cust_id in (456) to the update statement. Similarly, while the table is being deleted, only the rows for which the user is authorized are deleted.
    Attempting to insert a row containing data for which the user is not authorized results in an error message. For example, in this query, LARA is attempting to add a record to the CUSTOMER table for an account not under her purview: Next Steps
    READ more about
    VPD Oracle9i Supplied PL/SQL Packages and Types Reference
    Oracle Privacy Security Auditing
    SQL> insert into bank.customers
    2 values (789,'KIM PARK');
    insert into bank.customers
    ERROR at line 1:
    ORA-28115: policy with check option
    violation
    According to the ACCESS_POLICY table, Scott has SELECT privileges on account 789—no other privileges for any other AM are listed in the table.
    Using policies in conjunction with functions ensures authorized access to specific records of a table. The rules are applied regardless of how the table is accessed, whether through an application or directly through an ad hoc query tool, such as SQL*Plus. Users see only rows for which they have been authorized.
    Policies can be applied to multiple tables, and a single policy function can be used by any number of policies. Listing 6 shows a policy on an ACCOUNTS table that uses the get_sel_cust_id function initially created for use with the CUSTOMERS table.
    Arup Nanda ([email protected]) is the chief database architect at Proligence Solutions ( Printer View
    http://otn.oracle.com/oramag/oracle/04-mar/o24tech_security.html
    Joel Pérez
    http://otn.oracle.com/experts

  • Upgrade question from 10g to 11gr1

    Hi All
    Planning to upgrade 10.2.0.4.0 to 11.1.0.7.0
    After executing utlu111i.sql
    SQL> @utlu111i.sql
    Oracle Database 11.1 Pre-Upgrade Information Tool    07-19-2011 10:56:38
    Database:
    --> name:          TEST
    --> version:       10.2.0.4.0
    --> compatible:    10.2.0.1.0
    --> blocksize:     8192
    --> platform:      Linux x86 64-bit
    --> timezone file: V4
    -----Truncated-----------------
    Components: [The following database components will be upgraded or installed]
    --> Oracle Catalog Views         [upgrade]  VALID
    --> Oracle Packages and Types    [upgrade]  VALID
    --> JServer JAVA Virtual Machine [upgrade]  VALID
    --> Oracle XDK for Java          [upgrade]  VALID
    --> Oracle Workspace Manager     [upgrade]  VALID
    --> OLAP Analytic Workspace      [upgrade]  VALID
    --> OLAP Catalog                 [upgrade]  VALID
    --> EM Repository                [upgrade]  VALID
    --> Oracle Text                  [upgrade]  VALID
    --> Oracle XML Database          [upgrade]  VALID
    --> Oracle Java Packages         [upgrade]  VALID
    --> Oracle interMedia            [upgrade]  VALID
    --> Spatial                      [upgrade]  VALID
    --> Data Mining                  [upgrade]  VALID
    --> Expression Filter            [upgrade]  VALID
    --> Rule Manager                 [upgrade]  VALID
    --> Oracle OLAP API              [upgrade]  INVALID
    **********************************************************************My question is Oracle OLAP API is invalid. Shall i go ahead with the upgrade or do i need to make it valid before upgrade? if so how?
    Thanks
    Ak

    ok So now getting
    --> Oracle Packages and Types [upgrade] INVALID
    Components: [The following database components will be upgraded or installed]
    --> Oracle Catalog Views         [upgrade]  VALID
    --> Oracle Packages and Types    [upgrade]  INVALID
    --> JServer JAVA Virtual Machine [upgrade]  VALID
    --> Oracle XDK for Java          [upgrade]  VALID
    --> Oracle Workspace Manager     [upgrade]  VALID
    --> OLAP Analytic Workspace      [upgrade]  VALID
    --> OLAP Catalog                 [upgrade]  VALID
    --> EM Repository                [upgrade]  VALID
    --> Oracle Text                  [upgrade]  VALID
    --> Oracle XML Database          [upgrade]  VALID
    --> Oracle Java Packages         [upgrade]  VALID
    --> Oracle interMedia            [upgrade]  VALID
    --> Spatial                      [upgrade]  VALID
    --> Data Mining                  [upgrade]  VALID
    --> Expression Filter            [upgrade]  VALID
    --> Rule Manager                 [upgrade]  VALID
    --> Oracle OLAP API              [upgrade]  VALID
    **********************************************************************

  • Question: WebLogic 4.0 Clustering on Solaris 2.6

    I've read through several of the postings about Clustering Weblogic Servers
              but there are still a couple of points that aren't clear. We have a Java application
              that is going to use two Weblogic Application servers to go against a Sybase
              database. We decided to use two Single CPU Weblogic servers in a clustered
              configuration for load balancing and in the event one goes down or needs to
              be taken off line. Each server currently has two NIC's. One NIC is connected
              to the LAN where our client PC's are located and the second NIC is going to
              connect to an isolated Network for the Multicast communication and an NFS
              mount for the Weblogic servers. Eventually we might add a third NIC in each
              server that will connect to the network that the Database servers reside on but
              for now they will use the same NIC as the Client PC's.
              The part that isn't clear is the portion about the Multicast addresses. In some
              messages people try to assign and bind the Mutlicast address to the NIC and
              in others they just use a normal IP address and it looks as if the Mutlicast
              address is just configured in the Weblogic properties file. I would like to use
              the second NIC in the servers for the NFS mount and the Multicast communication
              between the servers. Is there anything special to configure this on a SUN solaris
              server? If you have any ideas please let me know.
              Regards,
              Robert
              

    There is a bug report that makes me think this is broken in Solaris 2.6, but
              in theory the following should work:
              Get rid of '/usr/sbin/route add -interface -netmask "240.0.0.0" "224.0.0.0"
              "$mcastif"' in /etc/init.d/inetsvc which sets the "standard" multicast
              address space starting at 224.0.0.0 with a mask of 240.0.0.0.
              Add to your taste:
              /usr/sbin/route add -interface -netmask "255.0.0.0" "235.0.0.0"
              "host1.foo.com"
              /usr/sbin/route add -interface -netmask "255.0.0.0" "236.0.0.0"
              "host2.foo.com"
              /usr/sbin/route add -interface -netmask "255.0.0.0" "237.0.0.0"
              "host3.foo.com"
              Where host1.foo.com, host2.foo.com, etc. are in your local hosts file and
              are set to the local IP address of the interface you want to bind each
              multicast class C address space to.
              netstat -r should give you the visual confirmation that things are
              configured correctly...
              Try snoop on a different machine to see if things are working on the wire:
              snoop -d hme0 multicast | grep -v ETHER
              Where hme0 is the ethernet interface in the proper Ethernet layer 2 domain
              that you want to analyze. Make sure you see the traffic you expect.
              Jim Hayes
              mailto://[email protected]
              Vinny Carpenter <[email protected]> wrote in message
              news:[email protected]...
              > Hi Robert. If you wish to keep the mutlicast traffic off your main
              network, why don't you just move the
              > WebLogic servers in their own network segment. We keep all of our
              WebLogic servers in their own network
              > segment and it works great. Hope this helps
              >
              > --Vinny
              >
              > "Robert L. Doerr" wrote:
              >
              > > I got a message from our Local Weblogic rep and he said (according to
              his inside support)
              > > that what I am trying to do can not be done. Apparently the Mutlicast
              traffic has to use the
              > > same NIC that the clients use to access the WebLogic Server. There is
              no way to use an
              > > isolated LAN for the Mutlicast packets like I originally wanted to do.
              I can still use the second
              > > NIC to go directly to the Database servers on the back end though. If
              anyone does know of
              > > a way to make the Muticast packets use a specific NIC please let me
              know.
              > >
              > > Regards,
              > >
              > > Robert
              > >
              > > "Robert L. Doerr" wrote:
              > >
              > > > Do you mean that the WLS looks for Multicast boradcasts on the same
              port it expects to
              > > > get requests from the clients? Without clustering enabled the WLS is
              listening on port
              > > > 7003 on the Primary NIC for communication with our Java application on
              the client. I thought
              > > > that the Muticast traffic was on a different port. If it uses the
              same port then how can we
              > > > tell the WLS to use the Primary NIC for communication and the second
              NIC strictly for
              > > > the Multicast traffic. I'm still not clear on this issue.
              > > >
              > > > Regards,
              > > >
              > > > Robert
              > > >
              > > > Sazi Temel wrote:
              > > >
              > > > > Hi Robert, you can use two or more NICs per server or you can bind
              multiple address to the same
              > > > > NIC. Every WLS server that is member of the cluster (regardless a
              server has single or multiple
              > > > > NICs, regardless if a WLS server has its own IP bind to its own NIC
              or a NIC used by multiple WLS
              > > > > servers) should listen the same port, you cannot have servers in a
              cluster listening different
              > > > > ports. Once the servers are configured being part of the cluster
              they will use the multicasting
              > > > > to communicate with each other. Note also that you should have
              license for the cluster
              > > > > configuration, you cannot combine multiple license files to make a
              cluster license file. Hope
              > > > > this will help.
              > > > >
              > > > > Regards,
              > > > > --Sazi
              > > > >
              > > > > "Robert L. Doerr" wrote:
              > > > >
              > > > > > I should clarify this a little better:
              > > > > >
              > > > > > Each Weblogic server has a NIC that it will use to communicate
              with all of the clients
              > > > > > on the LAN. The address of that NIC matches the one in the
              license file and it is
              > > > > > expecting requests to come in on port 7003. By default it looks
              as if the WebLogic
              > > > > > server looks for Multicasts on port 7001 of that same NIC. Since
              don't want that
              > > > > > Multicast traffic on the regular Network with the clients (It has
              no reason being there)
              > > > > > we want all the Multiport traffic to go in and out of another NIC
              with a different Network
              > > > > > address. Unless that second address is in the License file can we
              do this? I haven't
              > > > > > seen any messages or notes relating to this issue. It sounds like
              most people only
              > > > > > use one NIC per server.
              > > > > >
              > > > > > Regards,
              > > > > >
              > > > > > Robert
              > > > > >
              > > > > > "Robert L. Doerr" wrote:
              > > > > >
              > > > > > > Thanks, for the response. Can you control what NIC's that
              Multicast braodcasts
              > > > > > > will use? We would like to keep all of the Multicast traffic
              off the normal LAN and
              > > > > > > use the extra NIC's for this.
              > > > > > >
              > > > > > > Robert
              > > > > > >
              > > > > > > Sazi Temel wrote:
              > > > > > >
              > > > > > > > Your server should bind to a "normal" IP address... Use
              multicast address for clustered
              > > > > > > > servers communications, in most case you should do nothing for
              multicast address since if
              > > > > > > > you do not assign one WLS will use the default one
              (237.0.0.1).
              > > > > > > >
              > > > > > > > --Sazi
              > > > > > > >
              > > > > > > > "Robert L. Doerr" wrote:
              > > > > > > >
              > > > > > > > > I've read through several of the postings about Clustering
              Weblogic Servers
              > > > > > > > > but there are still a couple of points that aren't clear.
              We have a Java application
              > > > > > > > > that is going to use two Weblogic Application servers to go
              against a Sybase
              > > > > > > > > database. We decided to use two Single CPU Weblogic servers
              in a clustered
              > > > > > > > > configuration for load balancing and in the event one goes
              down or needs to
              > > > > > > > > be taken off line. Each server currently has two NIC's.
              One NIC is connected
              > > > > > > > > to the LAN where our client PC's are located and the second
              NIC is going to
              > > > > > > > > connect to an isolated Network for the Multicast
              communication and an NFS
              > > > > > > > > mount for the Weblogic servers. Eventually we might add a
              third NIC in each
              > > > > > > > > server that will connect to the network that the Database
              servers reside on but
              > > > > > > > > for now they will use the same NIC as the Client PC's.
              > > > > > > > >
              > > > > > > > > The part that isn't clear is the portion about the Multicast
              addresses. In some
              > > > > > > > > messages people try to assign and bind the Mutlicast address
              to the NIC and
              > > > > > > > > in others they just use a normal IP address and it looks as
              if the Mutlicast
              > > > > > > > > address is just configured in the Weblogic properties file.
              I would like to use
              > > > > > > > > the second NIC in the servers for the NFS mount and the
              Multicast communication
              > > > > > > > > between the servers. Is there anything special to configure
              this on a SUN solaris
              > > > > > > > > server? If you have any ideas please let me know.
              > > > > > > > >
              > > > > > > > > Regards,
              > > > > > > > >
              > > > > > > > > Robert
              > > > > > >
              > > > > > > --
              > > > > > > ------------------------------------------------------------
              > > > > > > Robert L. Doerr (MCNE, MCP, A+)
              > > > > > > 26308 Cubberness
              > > > > > > St. Clair Shores, MI 48081
              > > > > > > Tel: (810) 777-1313
              > > > > > > e-mail: [email protected]
              > > > > > > WEB Site: http://www.robotswanted.com
              > > > > > > "Keeping Personal Robots alive!"
              > > > > > > Heathkit HEROS (Jr, 1, & 2000), Androbots, & MAXX STEELE.
              > > > > > > ------------------------------------------------------------
              > > > > >
              > > > > > --
              > > > > > ------------------------------------------------------------
              > > > > > Robert L. Doerr (MCNE, MCP, A+)
              > > > > > 26308 Cubberness
              > > > > > St. Clair Shores, MI 48081
              > > > > > Tel: (810) 777-1313
              > > > > > e-mail: [email protected]
              > > > > > WEB Site: http://www.robotswanted.com
              > > > > > "Keeping Personal Robots alive!"
              > > > > > Heathkit HEROS (Jr, 1, & 2000), Androbots, & MAXX STEELE.
              > > > > > ------------------------------------------------------------
              > > >
              > > > --
              > > > ------------------------------------------------------------
              > > > Robert L. Doerr (MCNE, MCP, A+)
              > > > 26308 Cubberness
              > > > St. Clair Shores, MI 48081
              > > > Tel: (810) 777-1313
              > > > e-mail: [email protected]
              > > > WEB Site: http://www.robotswanted.com
              > > > "Keeping Personal Robots alive!"
              > > > Heathkit HEROS (Jr, 1, & 2000), Androbots, & MAXX STEELE.
              > > > ------------------------------------------------------------
              > >
              > > --
              > > ------------------------------------------------------------
              > > Robert L. Doerr (MCNE, MCP, A+)
              > > 26308 Cubberness
              > > St. Clair Shores, MI 48081
              > > Tel: (810) 777-1313
              > > e-mail: [email protected]
              > > WEB Site: http://www.robotswanted.com
              > > "Keeping Personal Robots alive!"
              > > Heathkit HEROS (Jr, 1, & 2000), Androbots, & MAXX STEELE.
              > > ------------------------------------------------------------
              >
              

  • Have few questions to ask regarding clusters

              1. In Weblogic server, how many servers can we scale to?(upper and lower limits,limitations)
              2.How do you provide load balancing across database connections and allow RDBMS
              load balancing for read only access?
              

    1. In Weblogic server, how many servers can we scale to?(upper and lower          limits,limitations)
              Depends on the app. Some will scale easily to 30 or 40 servers. Some won't
              scale to 2.
              > 2.How do you provide load balancing across database connections and allow
              RDBMS
              > load balancing for read only access?
              Depends. We use a driver that sends different users to different databases
              for some applications. Also we provide Coherence (http://www.tangosol.com/)
              to cache in the app tier with clustered apps.
              Peace,
              Cameron Purdy
              Tangosol, Inc.
              Clustering Weblogic? You're either using Coherence, or you should be!
              Download a Tangosol Coherence eval today at http://www.tangosol.com/
              "Amol" <[email protected]> wrote in message
              news:3c6cd4d8$[email protected]..
              > ...
              

  • Beginning Design question - Forms 10g

    This is the first form I have been asked to develop. I am designing a forms 10G application. It is to be a read only interface to multiple history tables in a 10 R2 database. I want to be able to have the user select a year and then populate the form with that year's data. Will I have to create a form for each year? or is there a way to pass the year as a parameter and then select a datablock built from each table and associate that datablock with just 1 form. OR, if that is a lousy design, what is the best approach?

    Hi,
    You just need one form, but some datablocks (maybe).
    If you have a database table for each 'year' then you need a datablock for each year.
    Create a stacked canvas for each datablock.
    To change beetwen blocks use:
    go_block('block_name');If you have just one table then you can use set_block_property
    set_block_property('block_name', default_where, 'where clause');
    Hope it helps!
    Edited by: Ruddy Guerra on 14-abr-2011 21:03

  • Question about 10g EM w/ 2 databases

    I've got 2 databases on one machine. How is EM supposed to be configured properly to see both of them?
    Right now I've got 2 directories under $ORACLE_HOME/<hostname>_<SID>, one per SID.

    Add the service names to OEM. You have to look for that option.
    Joel Pérez

  • 10g Personal Edition Question

    I am newbie and have a question regarding 10g personal edition. I see that this version supports one user connection but I am trying to find out if the user connection can be remote (from another workstation) or if it has to be local.
    I have (2) XP boxes and want to install 10g personal edition on one XP box then access the DB from the other XP box. Can this be done?
    Thanks in advance for any input!

    Like the original poster, I too am a total Oracle newbie, considering migrating to Oracle from the outstandling but comparatively feature-poor PostgreSQL, and I'm trying the downloadable version of the Personal Edition database 10gR2, and I cannot for the newbie life of me figure out how to use this thing remotely, so that I can have Oracle running on my server at home and use it/develop with it using Enterprise Manager and SQL Developer from, say, a coffeeshop or other remote location via the Internet. (Exception: remote control software--RADMIN works well--but this is not a good non-emergency solution.)
    Ultimately, it would also need to be secure (SSL or something), but at this point, I cannot get it to work at all. I know how port-forwarding and other firewallish things work, but I suspect the problem is that the Personal Edition uses a local loopback connection which may be inaccessible from the Net?
    Of course I'm perfectly willing to buy the product if I can be assured that this will ultimately work reliably--is that's what's necessary to get it to work (the downloadewd version being crippled somehow?), or to get any support (folks say Metalink is helpful)?
    Usually, the documentation is all I need to make something go, but the Oracle documentation I've looked at so far is at once incredibly expansive and not very helpful to me. (I may not know enough for it to be helpful: most of the documentation seems to assume you're already fully immersed in the Oracle world, and little of it seems specific to the Personal Edition's quirks.)
    Any pointers to explicit documentation, let alone some step-by-step "here's what you have to do" instructions, would be tremendously appreciated.
    Thanks!

  • Clustering instances

    bit unexperienceed with the 10g (9.0.4) app server.
    2 questions.
    1. Clustering, I am going to cluster my app servers for high availabaility. I know i can deploy my app once within one app server and it will be deployed across all others. Now my app will be in a specific oc4j instance, do I need to create the instance in all app servers myself or when an app server joins the cluster will it be created because the baseline configuration has it in?.
    2. I know its possible to precompile jsp pages before they are used in the app server, however I cannot find the setting where I can do this (or if it is advisable)
    (http://asktom.oracle.com/pls/ask/f?p=4950:8:::::F4950_P8_DISPLAYID:23862629497693) is where I found out you can
    Cheers
    Dave

    Yes.
              "Bala K" <[email protected]> wrote in message
              news:3c47359a$[email protected]..
              > Can I have the following configuration?
              > Two cpus each have 2 instances, one instance
              > listening on port p1 and another on p2.
              > Can I have 2 clusters one with instances listening on p1 and another with
              instances listening on p2?
              

  • Multiple clusters on a single machine

    Hi everyone. I have a question about running 2 clusters on a same
              machine.
              Here's the scenario. We have 8 WebLogic server (4.5.1 SP4) instances
              running on
              2 servers (Sun Enterprise 5500/Solaris 2.62) right now for one of our
              eCommerce
              application.
              Now we are getting ready to deploy a 2nd application and we want it to
              have its
              own cluster. So we are hoping to run 4 more WebLogic servers on each of
              the 2
              machines. So machine #1 would have 4 WebLogic servers for cluster A and
              4
              WebLogic servers for cluster B. And machine #2 will have 4 WebLogic
              servers for
              cluster A and 4 WebLogic servers for cluster B.
              All of these servers have unique IP address and each cluster has a
              different
              name and multicast IP address. Both clusters are using in-memory
              replication.
              On both machines, each of the cluster member will share the same global
              weblogic.properties file. The cluster specific portion resides in the
              cluster
              directory.
              Is anyone currently running such a configuration?? Is there something
              we should
              worry about?? Since both of these servers are in their own isolate
              network
              segment, is there any danger of multicast broadcast bleeding into the
              different
              cluster?? Thanks in advance.
              

    Hi Michael and thanks.. No, this is 2 systems on the same network and not two
              independent networks. Here's what I found that's very strange. I started
              server 1 ->4 for the first cluster and everything worked fine. Now I started
              server 5 on the first machine that's in a different cluster and has a totally
              different multicast IP address and yet, server 1 of the first cluster recorded
              server 5 joining the cluster. Server 2, 3 and 4 of the first cluster did not
              have RJVM 'Creating connection' message in there.
              On server 5's startup, server 1 decided to make server 5 it's secondary pair and
              I get the following error as server 5 does not have the same code for cluster1.
              After that error, I started getting tons of
              weblogic.cluster.replication.BadStatusException error in the logs for server 1
              listed below. I'm sure running multiple clusters off an enterprise level
              server is something people want to do. Is this a known issue?? Is this a bug
              in the server or the VM?
              Our environment: Sun 5500 Enterprise servers running Solaris 2.62, WebLogic
              4.5.1 SP4.
              Sun Jan 09 15:19:40 CST 2000:<I> <RJVM> Creating connection to
              stingray05.xx.com/10.x.x.x t3:7001 http:7001 t3s:7002 https:7002
              -1353087187406741800
              Sun Jan 09 15:19:41 CST 2000:<E> <MulticastSocket> Multicast socket receive
              error:
              Sun Jan 09 15:19:41 CST 2000:<E> <MulticastSocket>
              java.lang.ClassNotFoundException: Could not instantiate new instance of
              [com.strong.brokerage.server.CustomerSessionBeanHomeImpl_WLStub]
              Either the class is missing from classpath, the class does not have a public
              constructor, or the constructor raised an error; the actual exception was a :
              java.lang.NoClassDefFoundError: com.strong.brokerage.common.util.RegistrationCtx
              at java.lang.Throwable.<init>(Compiled Code)
              at java.lang.Exception.<init>(Compiled Code)
              at java.lang.ClassNotFoundException.<init>(Compiled Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readLeftover(Compiled Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readObjectBody(Compiled Code)
              at weblogic.common.internal.WLObjectInputStreamBase.readObject(Compiled
              Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readObjectWL(Compiled Code)
              at weblogic.service.ReplicaAggregate.readExternal(Compiled Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readPublicSerializable(Compiled
              Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readLeftover(Compiled Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readObjectBody(Compiled Code)
              at weblogic.common.internal.WLObjectInputStreamBase.readObject(Compiled
              Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readObjectWL(Compiled Code)
              at weblogic.cluster.BasicServiceOffer.readObject(Compiled Code)
              at weblogic.cluster.AnnotatedServiceOffer.readObject(Compiled Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readPublicSerializable(Compiled
              Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readLeftover(Compiled Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readObjectBody(Compiled Code)
              at weblogic.common.internal.WLObjectInputStreamBase.readObject(Compiled
              Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readObjectWL(Compiled Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readArrayList(Compiled Code)
              at weblogic.cluster.Announcement.readObject(Compiled Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readPublicSerializable(Compiled
              Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readLeftover(Compiled Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readObjectBody(Compiled Code)
              at weblogic.common.internal.WLObjectInputStreamBase.readObject(Compiled
              Code)
              at
              weblogic.common.internal.WLObjectInputStreamBase.readObjectWL(Compiled Code)
              at weblogic.cluster.TMSocket.execute(Compiled Code)
              at weblogic.t3.srvr.ExecuteThread.run(Compiled Code)
              Sun Jan 09 15:24:42 CST 2000:<E> <HTTP> Servlet request terminiated with Error:
              Sun Jan 09 15:24:42 CST 2000:<E> <HTTP>
              weblogic.cluster.replication.BadStatusException: update found
              -5050872662681587012 but it is not the secondary
              at weblogic.rmi.extensions.BasicRequest.sendReceive(Compiled Code)
              at
              weblogic.cluster.replication.ReplicationManager_WLStub.update(Compiled Code)
              at
              weblogic.cluster.replication.ReplicationManager.updateSecondary(Compiled Code)
              at weblogic.servlet.internal.session.ReplicatedSession.sync(Compiled
              Code)
              at
              weblogic.servlet.internal.session.ReplicatedSessionContext.sync(Compiled Code)
              at weblogic.servlet.internal.ServletRequestImpl.syncSession(Compiled
              Code)
              at weblogic.servlet.internal.ServletStubImpl.invokeServlet(Compiled
              Code)
              at weblogic.servlet.internal.ServletContextImpl.invokeServlet(Compiled
              Code)
              at weblogic.servlet.internal.ServletContextImpl.invokeServlet(Compiled
              Code)
              at weblogic.socket.MuxableSocketHTTP.invokeServlet(Compiled Code)
              at weblogic.socket.MuxableSocketHTTP.execute(Compiled Code)
              at weblogic.t3.srvr.ExecuteThread.run(Compiled Code)
              --------------- nested within: ------------------
              weblogic.utils.NestedError: Tried to update secondary, but it thought it was the
              primary
              - with nested exception:
              [weblogic.cluster.replication.BadStatusException: update found
              -5050872662681587012 but it is not the secondary]
              at java.lang.Throwable.<init>(Compiled Code)
              at java.lang.Error.<init>(Compiled Code)
              at weblogic.utils.NestedError.<init>(Compiled Code)
              at weblogic.servlet.internal.session.ReplicatedSession.sync(Compiled
              Code)
              at
              weblogic.servlet.internal.session.ReplicatedSessionContext.sync(Compiled Code)
              at weblogic.servlet.internal.ServletRequestImpl.syncSession(Compiled
              Code)
              at weblogic.servlet.internal.ServletStubImpl.invokeServlet(Compiled
              Code)
              at weblogic.servlet.internal.ServletContextImpl.invokeServlet(Compiled
              Code)
              at weblogic.servlet.internal.ServletContextImpl.invokeServlet(Compiled
              Code)
              at weblogic.socket.MuxableSocketHTTP.invokeServlet(Compiled Code)
              at weblogic.socket.MuxableSocketHTTP.execute(Compiled Code)
              at weblogic.t3.srvr.ExecuteThread.run(Compiled Code)
              Michael Girdley wrote:
              > Vinny,
              >
              > It sounds like the real hard problem here is a system administration and
              > configuration issue. If you have two independent networks with independent
              > systems, then you should not have a problem running two clusters across two
              > machines. You just need to make sure that your routing for multicast is
              > appropriately configured so that they don't confuse the two different
              > WebLogic server clusters.
              >
              > I hope this helps. I might also recommend contacting our sales organization
              > who can put you in touch with a systems engineer who is a professional at
              > making these types are recommendations.
              >
              > Thanks!
              >
              > --
              > Michael Girdley
              > BEA Product Manager
              >
              > Vinny Carpenter <[email protected]> wrote in message
              > news:[email protected]...
              > > Hi everyone. I have a question about running 2 clusters on a same
              > > machine.
              > > Here's the scenario. We have 8 WebLogic server (4.5.1 SP4) instances
              > > running on
              > > 2 servers (Sun Enterprise 5500/Solaris 2.62) right now for one of our
              > > eCommerce
              > > application.
              > >
              > > Now we are getting ready to deploy a 2nd application and we want it to
              > > have its
              > > own cluster. So we are hoping to run 4 more WebLogic servers on each of
              > > the 2
              > > machines. So machine #1 would have 4 WebLogic servers for cluster A and
              > > 4
              > > WebLogic servers for cluster B. And machine #2 will have 4 WebLogic
              > > servers for
              > > cluster A and 4 WebLogic servers for cluster B.
              > >
              > > All of these servers have unique IP address and each cluster has a
              > > different
              > > name and multicast IP address. Both clusters are using in-memory
              > > replication.
              > > On both machines, each of the cluster member will share the same global
              > > weblogic.properties file. The cluster specific portion resides in the
              > > cluster
              > > directory.
              > >
              > > Is anyone currently running such a configuration?? Is there something
              > > we should
              > > worry about?? Since both of these servers are in their own isolate
              > > network
              > > segment, is there any danger of multicast broadcast bleeding into the
              > > different
              > > cluster?? Thanks in advance.
              > >
              > >
              > >
              > >
              

  • Oracle 10g on SLES9 AMD64

    Hi All,
    i have tried to set up a system SLES 9 AMD64 for testing Oracle 10g for AMD 64 .
    After installing the operating system, I added this package
    rpm -Uvh orarun-1.8-109.5.i586.rpm
    and enabled the user oracle.
    I you login as user oracle, the following error message is displayed :
    error while loading shared libraries : librt.so.1 : cannot open shared object files : no such fileSuse support answers to this question : Oracle 10g is certified on SLES9, but the orarun package is only supported for SLES9 i386 ?
    This is correct, but did not help me to continue my installation.
    Any ideas ?
    Best regards
    Werner

    Try: comment the following line in /etc/init.d/oracle and /etc/profile.d/oracle.sh:
    #test -d /lib/i686 && export LD_ASSUME_KERNEL=2.2.5
    Post your question on [email protected] for faster response :)
    You may find http://www.gesinet.it/oracle/oracle10gonsles9amd64.html useful.
    -Arun

Maybe you are looking for