Apex and remote database links.... Good idea or bad?

We would like to have one single Apex installation that would be used to develop many applications against many remote databases.
Database links would seem to be the obvious mechanism to achieve this goal, but we have had many issues with db links.
Also, our research indicates this might not be a good idea for various reasons.
Any opinions or experiences around this topic would be greatly appreciated.

we have had many issues with db links. In that case it's a bad idea.
It's not really an Apex question. If you can't get db links to work, it's a database issue. If it's performance over the db link, again it's a database issue (assuming modern networks aren't the bottleneck here). Remote views (instead of selecting from remote tables) often increase performance.
If the applications don't have to share data over the db links, you can set up one master application and several sub-applications on the different databases. The master is used to log in once. See:
http://www.oracle.com/technetwork/issue-archive/2009/09-may/o39security-101079.html

Similar Messages

  • HELP! FOR LOOP TO SCROLL THROUGH TABLE AND CREATE DATABASE LINK

    Hi,
    Here's the scenario, not much of a PL programmer, just basic SQL so really need some help people!
    I have 2 tables. 1 contains list of DB's and the other contains rules to follow.
    I need to create a loop that goes through the table containing the DB's and on each row a DB link is created (Only 1 link allowed!)
    Once created, the schema currently logged in with also has an account on the linked DB in order to run scripts- The scripts are stored centrally hence the requirement for the link to the target DB.
    There are numerous scripts that need to be executed and can all be called from 1 script, once executed the loop exists and the database link needs to be dropped.
    Once dropped, the first loop continues, creating a DB link for the next DB listed in the table (and all the scripts are fired again)
    This continues against all the DB's listed in the table.

    Hi BlueShadow,
    Thanks again for the response, you've hit the nail on the head. SQL scripts on a unix server, a loop goes through a table 1 at a time. Each row gets a link created and then all the scripts stored on the server are executed against the db linked to. So I'm assuming this is a loop within a loop.
    1 loop to go through the table to create the link and then another loop within once connected to execute all the scripts against the connected DB. Once the scripts are run, the loop exits and moves onto the next server and so on until all the servers have the scripts are run.
    It's PL/SQL scripts we're after and not shell scripts as this would free us from the OS constraints.
    We have to drop the links due to security. Any idea on o

  • Where can I find one of those silicone body covers (not hardshell) for the Mac Book Pro? And is this a good idea?

    Where can I find one of those soft silicone body covers (not hardshell) for the Mac Book Pro? And is this a good idea since Apple no longer makes them? I-phones use them now.

    Personally I use the Speck hard case on both of my MacBooks and they are the greates! Look around on Amazon and eBay and you can find them for $30-40. I wouldn't really recomend a silicon case for a macbook, but it's your preference.

  • VLD-1124: Schema and/or Database link physical configuration parameters

    Hi!
    The warning message:
    VLD-1124: Schema and/or Database link physical configuration parameters should not be set for Table1.
    Is it possible to eliminate this warning message ? Anyway , Thanks for your response!
    Bye!

    Hi guys,
    I am facing a strange problem. The code generated for my mapping has NOT used the dblinks. My OWB version is 10.1.0.4
    This is my problem in brief.
    I have installed OWB newly and started to do a sample task. I created a very simple one to one table population mapping from source to the target schema. Both are in the same database.
    When i generated the code for the mapping, it gave a warning 'VLD-2771: System privileges may not allow extraction from source EMP'.
    When i checked the code which was generated, i couldnot see any dblinks associated with my source table( which seems strange)
    CURSOR "INGRP_c" IS
    SELECT
    "EMP_SRC_TRG_CONN"."EMPNO" "EMPNO",
    "EMP_SRC_TRG_CONN"."ENAME" "ENAME",
    "EMP_SRC_TRG_CONN"."JOB" "JOB",
    "EMP_SRC_TRG_CONN"."MGR" "MGR",
    "EMP_SRC_TRG_CONN"."HIREDATE" "HIREDATE",
    "EMP_SRC_TRG_CONN"."SAL" "SAL",
    "EMP_SRC_TRG_CONN"."COMM" "COMM",
    "DEPT_SRC_TRG_CONN"."DEPTNO" "DEPTNO_1",
    "DEPT_SRC_TRG_CONN"."DNAME" "DNAME",
    "EMP_SRC_TRG_CONN"."DEPTNO" "DEPTNO",
    "DEPT_SRC_TRG_CONN"."LOC" "LOC"
    FROM "SCOTT"."EMP" "EMP_SRC_TRG_CONN" ;
    In brief this is the process i have done.
    Source schema : SCOTT & Target schema : TRG_SCHEMA
    1) I have created source (for SCOTT) and target(for TRG_SCHEMA) modules.
    2) I have also created DBLinks, Locations and Connector from source to the target locations.
    3) I registered both the source location and target locations.
    4) Validated, Generated and Deployed the Connector from source to the target.
    I was unable to trace the error. Did i miss anything in the configuration? or during the installation of OWB.
    Though it is a very old post, i hope someone can help me out here.
    Thanks in Advance,
    Sri

  • ORACLE 11g + PHP5 problem: "fetch out of sequence" on remote database link

    Hi!
    I have a new server with oracle 11g (11.1.0.7) database and apache/php5 (actual build) on linux.
    connection type is dedicated.
    when doing a connection to a remote AS/400 database I always get this error:
    "Warning: oci_fetch_array() http://function.oci-fetch-array: ORA-01002: fetch out of sequence ORA-02063: preceding line from..."
    when doing the same simple "select * from database link" at my old server with oracle 9i and apache/php5 no problem occurrs.
    this happens at AS/400 database link only. normal oracle database links work fine.
    Anyone an idea how to solve this problem or where the problem is????
    further info:
    when doing " select * from [ database link ] where rownum < 11 " it works, but when doing the query with more than 10 results I get the error. Any idea?
    bye,
    Oliver
    Edited by: user501548 on 10.10.2008 02:20

    well a fetch out of sequence usually indicates that a cursor has been closed before the process has finished fetching or some such thing.
    I'm not aware of there being a specific problem with the situation you outline, but maybe if you could provide all the connection setup and how you are trying to connect that may give a better idea.
    Also, you may be better asking on the General Database Discussions forum as this isn't really a SQL or PL/SQL problem

  • Creating a hierarchy based on a remote database link

    I am trying to create a hierarchy based on a synonym for a remote database table. The hierarchy wizard works correctly, and the table columns do appear as choices in the drop boxes. However, when I try to run the application, I get the error "No data found. This data does not exist or you do not have privileges to view it". Does anyone know how to resolve this error? Or how to work around this?
    I can successfully create and query a form based on the remote table. And, I am able to create a hierarchy from other tables (which do not use a database link) without a problem.
    Thanks for your help!

    This is the query which I am using to build a report
    SELECT
    rollup_timestamp "Date",
    max(decode(target_guid,'199F0B201A3D71A63040BADFAA4F9E90',average,0)) host1,
    max(decode(target_guid,'3FB1329F59339C07E11304B69DC4E594',average,0)) host2
    FROM "sysman.MGMT$METRIC_DAILY"
    WHERE
    (target_guid='199F0B201A3D71A63040BADFAA4F9E90'
    or
    target_guid='3FB1329F59339C07E11304B69DC4E594')
    AND
    metric_name='Load'
    AND
    metric_column='memUsedPct'
    AND
    rollup_timestamp >= to_date('01-10-2009','dd-mm-yyyy') and rollup_timestamp <= sysdate
    GROUP BY rollup_timestamp
    ORDER BY "Date"
    And for your previous question I couldn't even create a page.
    Regards
    Balaji
    Edited by: user7290747 on 6/01/2010 16:17

  • Use of "Rel" Attribute in Links -- Good Idea?

    Hi --
    I was reading about various SEO methods recently and came
    across several
    pages which advocate the use of the "Rel" attribute on a
    hyperlink. The idea
    being that the "Rel" may influence how a search engine views
    a link and
    possibly help with ranking. For example: <a
    href="help.html" rel="Help">
    need help?</a>.
    Sounds like a good idea, but the more I read the confused I
    became. A Danny
    Sullivan article about use of the attribute "nofollow" used
    by major search
    engines made sense. It was the other article I read which
    used all
    different types of attributes that confused me such as
    rel="met colleague",
    rel="alternate". Is there someplace I can go to find a list
    of widely
    accepted attributes?
    Anyone have any experience with this?
    John

    Thanks, Joe. Yep, the source is the best place to start.
    Should have
    thought of that myself.
    I suspect you're assessment is correct...won't hurt, might
    help.
    Thanks again.
    John
    "Joe Makowiec" <[email protected]> wrote in
    message
    news:[email protected]..
    > On 02 Apr 2008 in macromedia.dreamweaver, Tarvardian
    wrote:
    >
    >> Is there someplace I can go to find a list of widely
    >> accepted attributes?
    >
    > How 'bout the source?
    >
    >
    http://www.w3.org/TR/html401/struct/links.html#adef-rel
    >
    > Note that it has a set of defined values:
    >
    >
    http://www.w3.org/TR/html401/types.html#type-links
    >
    > In looking over the documentation, and given my
    understanding of the way
    > things work in the here and now, I suspect that it's one
    of those "Won't
    > hurt, might help" kind of things.
    >
    > --
    > Joe Makowiec
    >
    http://makowiec.net/
    > Email:
    http://makowiec.net/contact.php

  • Remote Database Link

    I have to Database set up in 2 different machine. All are having rel 9.2.0.1.0 in them. I am trying to use private fixed database link to connect form one to another(from a 3rd client machine).
    say my orcale databases are:
    service_name for target database: orcl_tgt.x.y
    service_name for source database: orcl_source.p.q
    when i create a database link in orcl_target(orcl_tgt.x.y) from a 3rd machine(9i client installed) like this:
         CREATE DATABASE LINK "L1" CONNECT TO "user_name"
         IDENTIFIED BY "pswd1"
         USING 'orcl_source.p.q'
    it says link created ,but when i try to view data using the link from target
    select * from xyz@L1 it says
    ERROR at line 1:
    ORA-12154: TNS:could not resolve service name
    in our system global_names parameter is set to false in both the databases.
    Can anybody suggests how to solve this?

    justin is right
    first check your tnsname entry
    that it will work property
    for that first u connect to remote machine by sqlplus
    and use tnsname 'orcl_source.p.q'
    i think u r not created 'orcl_source.p.q' in tnsname files--------------------------
    Cause: The service name specified is not defined correctly in the TNSNAMES.ORA file.
    Action: Make the following checks and correct the error:
    Verify that a TNSNAMES.ORA file exists and is in the proper place and accessible. See the operating system specific manual for details on the required name and location.
    Check to see that the service name exists in one of the TNSNAMES.ORA files and add it if necessary.
    Make sure there are no syntax errors anywhere in the file. Particularly look for unmatched parentheses or stray characters. Any error in a TNSNAMES.ORA file makes it unusable. See the Oracle Net Services Administrator's Guide. If possible, regenerate the configuration files using the Oracle Network Manager.
    hope it will help you
    kuljeet pal singh

  • FPN and Remote Delta Link

    Hello,
    We are implementing FPN. We have a consumer portal in NW 7.0 SP15 and a BI-Java producer portal in NW 7.0 SP17.
    I would like to know whether it is necessary to have both consumer and producer portal to have the same SP stack in order to use Remote Delta Link scenario ?
    Thank you in advance.
    The-Hung Nguyen

    Hi,
    it's not necessary that the producer and consumer have the same SPS. That's one of the advantages of an FPN. Your portals can have different release cycles and SLA's.
    Sure, it's good to have both on the same SPS, but it's not necessary. Some functionality won't work, depending on the SPS of the Consumer / Producer, like RAI for BI if the SPS for the BI system simply doesn't know what RAI is (Remote Application Integration).
    br,
    Tobias

  • What is the lifespan of a Macbook Notebook Hard Drive, and is it a good idea to replace them more then once?

    In specific, I own a 2007 Macbook Notebook, and it is an Intel Core 2 Duo MB21.00A5.B07. Last year, it became necessary to replace the hard drive, and was informed that the reason for the replacement was that the hard drive had reached the end of its lifespan.
    My question is how long do Macbook notebook hard drives typically last? This also is in the context of frequent use of the notebook.
    Additionally, is it a good idea to replace the hard drive more then once, or will the hard ware on the notebook, after I'm assuming about six or so years, be too out of date to accommodate a second or third hard drive replacement?
    Also, since the replacement was still covered under an applecare plan (it's expired now), the general cost of replacing a Macbook notebook hard drive is not known? What is the cost generally expected of a hard drive replacement? This is in relation to repairs conducted at an applestore location (as opposed to a third party repair center). 

    justinfromlos angeles wrote:
    Also, this means that, the settings of a system like Snow Leopard, in addition to all other personal settings from a current hard drive, can be copied to and from an external hard drive?
    Your entire system can be copied to an external HD in bootable form. It's generally referred to as cloning, and there are a number of applications available to assist in the task.
    SuperDuper
    CarbonCopyCloner
    LaCie Silverkeeper
    To name but three.
    Furthermore, once cloned, should you have to replace the internal drive, it's easy to boot from the external and clone the system back to the internal.
    justinfromlos angeles wrote:
    If so, do you know where one can purchase an external hard drive and a replacement hard drive?
    As I don't live in the US, it's difficult for me to reccommend retailers, but any computer store (including Apple) can supply external HDs. Be aware that not all HDs from some manufacturers will play well with Macs; Western Digital I believe needs careful selection of the correct model. There are plenty of contributors who can no doubt help on that front. Personally I use LaCie and Iomega. Note that these are purpose built externals and come with their own enclosure and all cables reqd.
    One thing you will need to do with most of them is to use Disk Utility to partiton them correctly (GUID system for Intel Macs) and set the file system to Mac Extended (journaled) before cloning.
    justinfromlos angeles wrote:
    There is one last question. Say I've replaced the hard drive on the same computer for nine years or so. Despite the hard drive being new, would the hardware of the computer be too out of date to accept an up to date replacement hard drive?
    The hardware of the computer dictates what sort of internal drive you buy; earlier models used IDE interfaces, current models use SATA and these are not interchangable. However System Profiler will tell you what sort you have. It's still possible to buy compatible HDs for the early iMacs, c. 1998/99 so there's no reason to believe your model will suddenly become defunct.

  • Weblogic 5.1, Distributed Transaction and Oracle Database Link

    We are planning to add a new functionality to run in Weblogic server and it
    needs to update tables in two different Oracle databases. Since we are using
    Weblogic 5.1 and it is unrealistic to upgrade to 6.1 in short term, we have
    to deal with the fact that distributed transaction is not supported by JTS
    on 5.1.
    One workaround we can think of is to utilize Oracle database link. Instead
    of having two connection pools, the application talks to only one database
    and access the tables on the other ones through Oracle database link.
    Therefore, when a session bean tries to update two tables on database 1 in
    one transaction, one of the tables is just a link to the other database. We
    hope that even though distributed transaction is not supported by JTS in
    Weblogic 5.1, but since it is supported by Orable, it would work.
    Do you think the above workaround would work? Does anyone have experience on
    something like that?
    Thanks in advance,
    David Chen
    [email protected]

    Hi. You should be able to use any JDBC driver with WebLogic 5.1.
    It may be the JVM that complains, if you use too old a JVM...
    Joe Weinstein at BEA Systems

  • Appcelerator Titanium? good idea? bad idea?

    So in my attempt at learning Objective C and the iPhone SDK to build my own apps i have been directed to The Appcelerator Titanium program for Mac. Is this thing a good idea? It uses Javascript from what i understand is a bit easier to get behind than Object C or Java (which id like to eventually do Android apps to).
    So any opinions on this? good? bad? any input would be great.
    Thanks

    My iPhone started acting "loopy" in March. Locked up while I was on the road; had to buy a Go phone. After a complete restore... well, it continued to act flakey: would randomly ask for SIM PIN, would randomly restart itself, would loose signal which it regained after restart. Trip to the Genius bar in Tampa: replaced the phone, then had AT&T replace the SIM card. Problems returned. Trip to Genius bar in Orlando: another phone but problems persisted. Call to Apple: conference call w/ 2 Apple techs, AT&T tech and decided to get another SIM card from AT&T store but not turn on SIM PIN. Problems persisted. Another call to Apple on Saturday, May 3. Apple tech Lynn asked "Do you have a belt case w/ a magnetic clasp?" I said "Yes." She said "Get rid of it." Her daughter had similar problems w/ iPod which were caused by case w/ magnetic clasp. So I stopped using the case w/ magnetic clasp and my iPhone has been rock solid ever since. Kudos to Lynn. Note to other Apple techs: That should be the 1st question you ask.

  • Visual Studio/ Oracle set up with both local and remote database

    Hi, I am developing a database application in Visual Studio 2010, and have installed Oracle Database 11g Express on the same pc - (Windows 7, 64bit).
    Now I am faced with what I need to install next: I've seen that Oracle Developer Tools for Visual Studio is required/recommended, but as it comes as part of the ODAC suite, it asks if I want to install the client or server version. Which should it be?
    And would the situation change if I also needed to connect to a remote (Oracle 10g) database? I'm keen to avoid having tnsnames files all over the place, if at all possible, but would I need a separate Oracle client installation in either case?
    Many thanks,
    Nick

    Hi,
    Are you actually going to drag and drop things from Server Explorer, and or use Sever Explorer to create/edit/debug stored procedures and such? If so, you'll need (32 bit) Oracle Developer Tools installed, and you'll need to install it into a separate home from the Express home.
    If you're just going to "hand code" ODP.NET, you can use the ODP.NET that installs in the Express home. Note though that in VS2010, projects default to compile target of "x86", so to run a 64 bit app with the 64 bit ODP/client (assuming you installed 64 bit db home), you'll need to change it to x64.
    You can connect from ODP.NET without tnsnames via the EZConnect version of the connect string.. ie,"data source=//serverhost:port/db_service_name; user id...etc.."
    You can use 11g client against a 10g database.
    Hope it helps,
    Greg

  • Local and Remote Database

    Hi Guys,
    I would like your suggestion on the following situation I
    want to try, currently we have a database on a web-sever to deliver
    content, add members and created pages for the CRM functions for
    our sales staff.
    We have specific requirements for the CRM application part of
    the system that requires me to design a number of forms for CRM
    users plus the database has other fuctionallity for other parts of
    the business.
    Due to these forms and additionaly functionallity
    requirements it is not practical to continue hosting the database
    and need to host this part of our site locally.
    Therefore I am considering if the suggestion below would
    work:
    1. Host the database locally on our own server allowing
    internal users full functionallity locally using designed access
    forms.
    2. Continue hosting the website via hosting providor
    3. Create a connection string from the hosted website to our
    server for the database
    Ultimately the sensible solution is to host both the database
    and website via our own server, but due to setup problems we are
    having currently the site can be flaky, But we need to switch to
    the local database to improve the CRM part of the business in the
    very immediate future.
    If anyone has any suggestions, advise or further information
    related to this I would be grateful for the feedback.
    Regards: Les
    SBS 2003, DW8, ASP, Access 2000 DB

    Hi,
    Are you actually going to drag and drop things from Server Explorer, and or use Sever Explorer to create/edit/debug stored procedures and such? If so, you'll need (32 bit) Oracle Developer Tools installed, and you'll need to install it into a separate home from the Express home.
    If you're just going to "hand code" ODP.NET, you can use the ODP.NET that installs in the Express home. Note though that in VS2010, projects default to compile target of "x86", so to run a 64 bit app with the 64 bit ODP/client (assuming you installed 64 bit db home), you'll need to change it to x64.
    You can connect from ODP.NET without tnsnames via the EZConnect version of the connect string.. ie,"data source=//serverhost:port/db_service_name; user id...etc.."
    You can use 11g client against a 10g database.
    Hope it helps,
    Greg

  • Apex and Ajax tutorial link

    Hi,
    I want to use Ajax in my Apex application. How can I use it? Can anyone guide me to proper link or howtos or any example?

    Try this link, you should get plenty of examples on those pages.
    http://www.google.com/search?hl=en&q=APEX%2BAJAX&btnG=Google+Search
    Alexandre.

Maybe you are looking for