Which is the best way of deploying JSP (mod_ose or OSE) ?

I need to deploy some jsps that are using servlet api 2.2 and i would like to know which is the best way of deploying these jsps?
Thanks
RJC

Speaking of JServ Vs OSE, I have some related questions on it. I would appreciate if someone can answer:
1.For a stateful servlet (assuming that we are using Oracle JVM), Will it not call the init() method of the servlet each time for a client (because each client will have his own Oracle JVM)? If so, isn't that expensive because it needs to initialize the servlet again?
Again, for the same servlet, Will it not have a servlet context for each and every client? Isn't that expensive? ( A servlet context by definition maintains state information for all instances of a web application within any single JVM)
2. If we have an application that has some portion of it which is stateless and some portion of it which is stateful, how do we make a choice JDK JVM vs Oracle JVM?
Can we (for the same application) use both the JDK JVM as well as Oracle JVM? If so, how does the JDK JVM talk with Oracle JVM?
3. Since Oracle JVM is useful for stateful applications only, does it mean that we cannot use it for Stateless session beans? (In an application we might have stateless session beans, stateful session beans as well as entity beans) Should we be using JDK JVM or Oracle JVM?
4. For Servlets too, we might have stateless servlets and stateful servlets within the same application. If so, can we use JDK JVM for stateless servlets and Oracle JVM for stateful servlets? If so, how does the JDK JVM talk with Oracle JVM?

Similar Messages

  • Which is the best way Applets or JSP

    Hi I am trying to develop a application to
    run on the internet .It involves few master detail table in each form and 10 Different functions.The forms needs to have LOVs and other validations.Which will be the best way to go.Total tables involved would be around
    40 and coulumns in these tables would be (15 - 40).
    Any suggestions
    jagan
    null

    I'll give you my input/opinion based on my own experience. Others may have additional considerations to add, or may want to disagree with some of my points:
    Benefits of JSPs:
    1. This is an evolving technology that a lot of people are now using. It is very powerful, and the performance is pretty good.
    2. It is much easier to deploy than an applet.
    3. It will continue to improve and be easier to work with, especially with our upcoming 3.2 release.
    4. Does not require a lot of Java experience.
    Drawbacks of JSPs:
    1. No GUI editor for laying out the user interface. You are using HTML code in a text editor, which is trickier to see how the UI will look as you go. Again, this will continue to improve over time, and our 3.2 release will make it a little easier to deal with the HTML and Java code.
    Benefits of Applets:
    1. There is a UI editor for designing the layout, but layout in Java can be tricky.
    Drawbacks of Applets:
    1. Deployment can be very error prone.
    2. Subject to security restrictions such as the database you connect to and the webserver the applet runs from must be on the same machine. You can work around this, but it complicates matters.
    3. Clients need to have the Java Plug-in installed with their browser.
    4. Performance is not great, especially for a lot of data going back and forth.
    5. Requires that the applet code is downloaded to the client, temporarily at least, at runtime. The more code you have in your applet, the longer it takes to download when the client invokes it.
    Bottom line is that Applets are fine for small tasks, and are preferable for 'intranet' use versus 'internet' use (outside).

  • What is the best way of deploying a jsp and bc4j aplication

    Hi
    I would like to know what is the best way of deploying a jsp and
    bc4j aplication in ias 9i.
    thanks in advanced
    rjc

    In the page I simply referenced the facescontext directly... no need for a custom servlet.
    public void createcookie() {
    FacesContext fc = FacesContext.getCurrentInstance();
    HttpServletResponse resp = (HttpServletResponse)fc.getExternalContext().getResponse();
    Cookie userCookie = new Cookie("cookiename", "cookievalue");
    userCookie.setMaxAge(-1);
    userCookie.setMaxAge(3600);
    resp.addCookie(userCookie);
    return null;
    public String readcookie() {
    FacesContext fc = FacesContext.getCurrentInstance();
    ExternalContext ec = fc.getExternalContext();
    Map cookiemap = ec.getRequestCookieMap();
    if (cookiemap != null) {
    Cookie cookie = (Cookie) cookiemap.get("cookievalue");
    return cookievalue;
    For reference I could not get getCookies() method of HttpServletRequest to work.. it would only return JSESSIONID.

  • What is the best way to deploy application that uses web services?

    Hi all,
    I'm having some problems figuring out the best way to deploy our app now that we've switched over to using web services.
    I'm fairly new to Java and web services. From what I understand, JNLP and WebStart are methods to deploy Java clients to users and not for creating war files and the like.
    Here's a little background:
    We have a large PowerBuilder/Oracle application. The db consists of over 500 tables and the client consists of several thousand PB components. A user creates "transactions" which contain a series of "sub-transactions" within. Most of the data is collected and stored locally in the client in a series of datastores. When the "finalize" happens, the records are validated and sent to the database.
    We are in the process of moving each of the subtransactions (currently in the PB client) into subPROCESSES on a java project. We are using the JAX-WS framework to develop the web services in Netbeans. These web services aren't much more than remote xml as the messaging technology. For writing data back to the database, we are using the Java Persistence API to function outside of an EJB container but will shortly be migrating to the Glassfish application server to use several of the EJB container frameworks including the EntityManager.
    We haven't attempted any type of deployment and are unsure of where to start. Any suggestions would be so helpful and appreciated!
    Thanks!
    Edited by: doubleEspresso on Jan 10, 2008 8:06 AM

    >
    I'm fairly new to Java and web services. From what I understand, JNLP and WebStart are methods to deploy Java clients to users and not for creating war files and the like. >Correct, while Java Web Start has 'web' in the name, it has little if anything to do with web applications - certainly not providing much toward their installation. It is for launching rich client GUI based (AWT, Swing, SWT..) applications onto the end-user's desktop.
    There are some parts of JWS that might seem peripherally useful to the installation of a web-app., but it is really not a 'good fit'.
    >
    ..Any suggestions would be so helpful and appreciated!>You might try the forums for the 'web tier' APIs.
    <http://forum.java.sun.com/category.jspa?categoryID=20>
    Or perhaps the forum 'Java Technologies for Web Services' (under 'enterprise technologies')
    <http://forum.java.sun.com/forum.jspa?forumID=331>
    This one in 'BigAdmin' seems particularly relevant, 'Set up and Deploy'
    <http://forum.java.sun.com/forum.jspa?forumID=550>

  • Which is the Best way to upload BP for 3+ million records??

    Hello Gurus,
                       we have 3+million records of data to be uploaded in to CRM coming from Informatica. which is the best way to upload the data in to CRM, which takes less time consumption and easy. Please help me.
    Thanks,
    Naresh.

    do with bapi BAPI_BUPA_FS_CREATE_FROM_DATA2

  • What is the best way to deploy/update custom security realm classes to WLS 6.0?

    From the WLS 6.0 console, I see that I can specify the Java class that
    implements my custom security realm but I am wondering what is the best way
    to deploy/update this code. I don't see a way to do this from the console.
    Does this mean that I have to manually copy the class files over that
    implement my custom security realm?

    Thanks Danut,
    A jar file seems to be a good way to package it up but it sounds like it
    still needs to be manually copied to each Weblogic server install directory
    post-installation and whenever it is updated. I thought it would be nice to
    be able to deploy/update the custom security realm by uploading it through
    the Console just as you can with web applications and EJBs.
    Brian
    "Danut Prisacaru" <[email protected]> wrote in message
    news:3aba2db0$[email protected]..
    You have to have your Custom Realm class in the class path. I usually havea
    jar file with all the Custom Realm classes and that jar I copy it in thelib
    folder. Then I modify "startWebLogic.cmd" and I add to the classpath
    ".\lib\CustomRealm.jar"
    set
    CLASSPATH=.;.\lib\weblogic_sp.jar;.\lib\weblogic.jar;.\lib\CustomRealm.jar;
    >
    Be aware that in order to have you custom realm besides creating thecustom
    realm using the console you also have to create a custom caching andchoose
    that one as your default caching realm.
    Here is how the security settings are looking in my "config.xml"
    <CustomRealm Name="CustomRealm"
    RealmClassName="Custom.appserver.weblogic.security.CustomRealm"/>
    <CachingRealm BasicRealm="CustomRealm" CacheCaseSensitive="true"
    Name="CustomCachingRealm"/>
    <Realm CachingRealm="CustomCachingRealm" FileRealm="wl_default_file_realm"
    Name="wl_default_realm"/>
    <FileRealm Name="wl_default_file_realm"/>
    <Security GuestDisabled="false"
    Name="mydomain" PasswordPolicy="wl_default_password_policy"
    Realm="wl_default_realm"/>
    Danut

  • The best way to deploy and redploy of a multiple SOA application

    Hi all,
    I'm looking for the best way to deploy and redeploy a multiple SOA application.
    1) The multiple SOA application contains many projects. Some projects depend on another projects.
    2) WebLogic is in productive mode.
    I wound like automaticly deploy and redeploy my multiple SOA application.
    Does anyone have any experience of this?
    Many thanks
    PG

    We use ant scripts to automatically deploy and redeploy multiple composites.
    The ant scripts gets shipped along with the SOA installation. The ant script that is used for the deployment is:
    ${oracle.home}/bin/ant-sca-deploy.xml
    You can also use the same for your deployment.

  • Which is the best way for posting a large number of records?

    I have around 12000 register to commit to dababase.
    Which is the best way for doing it?
    What depends on ?
    Nowadays I can't commit such a large number of register..The dabatase seems hanged!!!
    Thanks in advance

    Xavi wrote:
    Nowadays I can't commit such a large number of registerIt should be possible to insert tens of thousands of rows in a few seconds using an insert statement even with a complex query such as the all_objects view, and commit at the end.
    SQL> create table t as select * from all_objects where 0 = 1;
    Table created.
    Elapsed: 00:00:00.03
    SQL> insert into t select * from all_objects;
    32151 rows created.
    Elapsed: 00:00:09.01
    SQL> commit;
    Commit complete.
    Elapsed: 00:00:00.00
    I meant RECORDS instead of REGISTERS.Maybe that is where you are going wrong, records are for putting on turntables.

  • Which is the best way for a called function to identify caller class name.

    Which is the best way for a called function to identify the caller class name .
    1)Using sun.reflect.Reflection from called function
                    Class caller = Reflection.getCallerClass(2);
                    System.out.println("Caller Class Name ::"+caller.getName());2) Analyzing current threads stack trace from called function
                    StackTraceElement[] stElements=Thread.currentThread().getStackTrace();
                    System.out.println("Caller Class Name ::"+stElements[3].getClassName());Is there any alternate ways to achieve the same .Which is the best way ?
    Called function doesn’t have any arguments, I don’t want t pass any arguments from caller function to called function.
    Plz help.
    With kind regards
    Paul

    798185 wrote:
    Which is the best way for a called function to identify the caller class name .
    Is there any alternate ways to achieve the same.SecurityManager
        // 0 is the anonymous SecurityManager class
        // 1 is this class (also works in static context)
        // 2 is calling class
        static Class getClass(int i) {
            return new SecurityManager() {
                protected Class[] getClassContext() {
                    return super.getClassContext();
            }.getClassContext();

  • Filter or Restrict on? Which is the best way to put it?

    Hello Experts,
    If there is a Characteristic/Key figure on a report in BEx, and I right click on it and find another data element say, 0FISPER it I EDIT. On right click on 0FISPER I a variable.
    My question is this, do we way that the key figure is RESTRICTED ON 0FISPER
    or, key figure is FILTERED by 0FISPER
    When do we use Filter or Restrict on? Which is the best way to put it? How best is it read, please include the variable in your discussion.
    I will appreciate this clarification.
    e.g. Company code EDIT shows Company codes ZPPTcomp: with nodes  8000, 9000
    Thanks.

    Hi Amanda,
    as per my understanding,
    you are checking the structure field, you have selected edit then you are viewing a key figure and 0fiscper restricted by varaible. am I right?
    if this is the case this is local selection and the key figure is restricted by fiscper variable means the restriction value user can enter in selection screen. if you select 0fiscper and select restrict there you can see single values, value range and you can also see varaible tab. by using varaible tab you can restrict to variable value.
    hope this gives you some understanding.
    Sathya

  • Which is the best way to create a connection object

    hi,
    Eventhough i am not new to java I just wanted to know that which is the best way for creating sql connection object.
    Is it using the DriverManager.getConnection(..,..,..) or using a datasource and getting the connection.
    Which is more efficient or Is there any other way to create sql connection which is even more better.
    thanks in advance.
    pradeepbhargav

    AFAIK all connections, DataSource ones included are created by calling the DriverManager. A DataSource is a factory for Connection objects (from JavaDocs)
    If you're talking about managing connections then it depends on the application. A DataSource is great for apps that can support many clients or are multi-threaded, but may require a JNDI implementation to get to it.
    If however you have an single threaded app that hits the database in a serial fashion, then DataSource is probably an overkill.
    Dave

  • I am currently running my mac booc pro with the osx 10.5.8 and i want to update it. Which is the best way?

    I am currently running my mac booc pro with the osx 10.5.8 and i want to update it. Which is the best way?

      Snow Leopard Purchase
    Computer Compatibility - Lion
    Mountain Lion/Mavericks
    Check that your computer is compatible with Mountain Lion/Mavericks.
    To check the model number hold down the option/alt key, go to the Apple menu and select System Information.
    MacBook Pro (Mid/Late 2007 or newer) model number 3,1 or higher
    Your Mac needs:
    OS X v10.6.8 or OS X Lion already installed
    2 GB or more of memory (More is better - 4 GB minimum seems to be the consensus)
    8 GB or more of available space

  • What is the best way to deploy an application ??

    Hi,
    Please advise me how to automate in a best way to Deploy 10 different applications(.ear/.war)
    to 10 different Clusters(containing 40 Managed servers) in wls6.1......and that
    too in a Production environment.
    I know that we can use weblogic Console to deploy an application BUT if i have
    2 unix boxes and each box contains 4-4 clusters(each containing 16-20 managed
    servers) .....then i have to open up 2 Admin server consoles to deploy each of
    the respective applications and that too deploying these application to 8 clusters.....OR
    i can use deploy commad to do the same.
    AND if i have like 100 clusters .....then i would have to "deploy" 100 times ......
    is there any way that we can BEST AUTOMATE this process.
    Thanks a lot.
    -sangita

    Thanks Tommy. but i would like to know a very important point here: this is using
    the weblogic.utility that you just talkd about.
    Suppose iam having an ADMIN server that has say 10 clusters and 50 managed servers
    underneath to take care.
    1 cluster each = 5 managed server
    now iam deploying a simple .war file to say a single managed server, i have found
    out that it takes a looot of time to get the application deployed.
    what do you say about this situation ?
    what is the best approach ??????
    -sangita
    "Tomy Mathew" <[email protected]> wrote:
    >
    >
    To deploy the application from the command prompt in WebLogic 6.1 you
    use the
    following command
    $JAVA_HOME/bin/java -cp $WL_HOME/weblogic.jar weblogic.deploy -port $WEBLOGIC_PORT
    -host $WL_HOST_IP $WEBLOGIC_CMD $WEBLOGIC_PWD -component $WL_Server_NAME:$WL_Server_NAME
    $WL_APPLICATION_NAME YourApplication.war
    $WEBLOGIC_CMD sould be DEPLOY for the first time deployment and UPDATE
    for redeployment
    For deploying in a cluster you better create a shell script which call
    the above
    command with different cluster ip address as a parameter.
    Tomy
    "sangita" <[email protected]> wrote:
    Thanks Rob for your expertise advise. I would check to see wls8.1's
    ANT tasks.....for
    wls6.1, i would guess shell scripts to be a better option for us.
    thanks again.
    Rob Woollen <[email protected]> wrote:
    If you're using 8.1, then it's pretty nice to use ant for
    configuration/deployment/starting and stopping servers etc. There's
    a
    wlconfig task to do configuration, wldeploy to deploy applications,
    wlserver to start/stop servers.
    For 6.1, it's probably not a huge win to use ant. It's up to you.
    If
    you're more comfortable using shell scripts, that's certainly a
    reasonable route.
    -- Rob
    sangita wrote:
    Thanks Rob.
    But all the stuff that ANT can do for me FOR DEPLOYMENT (ONLY) .....ican do it
    from a simple shell script too.
    And i understand of ANT's cross-platform environment and usage of
    JAVA
    .....but
    can i use ANT to do something like CONFIGURING my wls6.1 managed
    servers/clusters/creating
    managed servers/db connection pools ......and all the other configurationthat
    i do from wls console(or manually) ....????
    and if yes, do you know of any examples, suggestions, ideas ?????
    basically, if i need to sell(?) ANT to my juniors ......what are
    the
    basic reasoning
    i can put forward....keeping in mind that these guys use wls6.1 &
    wls8.1......and
    these guys do CONFIGURATION and DEPLOYMENT only.
    sorry for asking, may be these silly questions.
    thanks, sangita
    Rob Woollen <[email protected]> wrote:
    sangita wrote:
    Thanks Rob.
    Rob, do you still consider or suggest using ANT for deploying say
    20
    applications
    on a 10 clustered environment.Yes
    do you see any advantages using [ANT + weblogic.deploy]..... over
    simply
    using
    weblogic.deploy utility ???Yes. ANT essentially gives you a cross-platform environment that
    allows
    extensibility through java. It's very powerful.
    -- Rob
    -sangita
    Rob Woollen <[email protected]> wrote:
    sangita wrote:
    Rob,
    I have found the answer for my question #1 ...this is the URL:
    http://edocs.bea.com/wls/docs81/admin_ref/ant_tasks.html
    iam still hoping to hear back from you for #2, #3
    thanks a ton.
    "sangita" <[email protected]> wrote:
    Thanks Rob.
    quick question:
    1) do you know of any example which illustrates the use of ANT
    wrapper....or
    please
    direct me to a URL where i can find some documentation about
    it.
    2) does this ANT wrapper has any advantages over using weblogic.Deployer
    utility
    ??If you're doing deployment via ant, then it's a bit more convenientto
    use <wldeploy> rather than invoking a raw <java> process, but
    fundamentally it's just a wrapper around weblogic.Deployer
    3) i would certainly check to see weblogic.deploy for wls6.1
    ....can
    i use ANT
    for deployment purpose in wls6.1 ?You'll just need to wrap it in a <java> task
    See:
    http://ant.apache.org/manual/CoreTasks/java.html
    -- Rob
    Rob Woollen <[email protected]> wrote:
    If you were using WLS 7.0 or later, you could use weblogic.Deployer.
    If
    you were using 8.1, you could use wldeploy which is an ant task
    wrapper
    for weblogic.Deployer.
    In WLS 6.1, you'll have to use weblogic.deploy which is a bit
    cumbersome, but it's still certainly possible to script deployment.
    -- Rob
    sangita wrote:
    Hi,
    Please advise me how to automate in a best way to Deploy
    10
    different
    applications(.ear/.war)
    to 10 different Clusters(containing 40 Managed servers) in
    wls6.1......and
    that
    too in a Production environment.
    I know that we can use weblogic Console to deploy an application
    BUT
    if i have
    2 unix boxes and each box contains 4-4 clusters(each containing
    16-20
    managed
    servers) .....then i have to open up 2 Admin server consoles
    to
    deploy
    each of
    the respective applications and that too deploying these applicationto 8 clusters.....OR
    i can use deploy commad to do the same.
    AND if i have like 100 clusters .....then i would have to "deploy"100 times ......
    is there any way that we can BEST AUTOMATE this process.
    Thanks a lot.
    -sangita

  • Which is the best way to parse nested XML

    Hello, I am just getting back into using ColdFusion and I am
    working on a project that needs to parse some XML.
    The XML is generated by a third party document.
    I figured out how to grab individual values and such, but I'm
    not sure what is the best way to grab a value in relation to where
    it is in the XML.
    Following is a rather simplified example of the XML
    I need to grab the value of 'Rich Text' in relation to the
    'ID name' value.
    &lt;project&gt;
    &lt;CT text&gt;
    &lt;Group&gt;
    &lt;id name = text1&gt;
    &lt;Rich Text&gt;some sample text&lt;/Rich
    Text&gt;
    &lt;/id&gt;
    &lt;/Group&gt;
    &lt;/CT Text&gt;
    &lt;CT text&gt;
    &lt;Group&gt;
    &lt;Group&gt;
    &lt;id name = text1&gt;
    &lt;Rich Text&gt;some sample text&lt;/Rich
    Text&gt;
    &lt;/id&gt;
    &lt;/Group&gt;
    &lt;/Group&gt;
    &lt;/CT Text&gt;
    &lt;CT text&gt;
    &lt;id name = text2&gt;
    &lt;Rich Text&gt;some more sample text&lt;/Rich
    Text&gt;
    &lt;/id&gt;
    &lt;/CT Text&gt;
    &lt;/project&gt;
    Once parsed, the info I need to output on the page would look
    like
    text 1
    some sample text
    text 2
    some more sample text
    The part that makes this difficult for me is that there may
    be 0 'Group' values or there might be a number of them. As such, I
    can't just grab the value by listing the tree out, since it changes
    depending on the document (there are thousands of these documents
    that I need to parse).
    From what I have seen in some XML examples, there are a
    number of ways to go about doing this, but before I put a lot of
    time into any of them, I was wondering what the 'best' way to do
    this is.
    XSLT? XMLSearch? Structures? Some combination of those or
    other functions?
    P.S. what little I know of using ColdFusion with XML all
    comes from the doc written by Nate Weiss.
    I was a bit let down that after spending well over $1,000 for
    ColdFusion, Adobe left out the only two books that were worth
    printing (the CFML reference guide and the Advanced book). It
    wouldn't have bothered me as much if there was something on the
    product page listing the manuals that said 'we only actually give
    you 2 of the ones listed'.
    Best regards
    Kevin R.

    Hello,
    Could someone do me a huge favor and provide some sample code
    that would parse/output the above xml so that the output would look
    something like the following. . .
    text 1 = some sample text
    text 2 = some more sample text
    The problem I am having is displaying the xmltext for the
    RICH TEXT element in relation to the ID element.
    The code below will return the value of the RICHTEXT
    (incorrectly named in the original XML examples) easily enough, but
    I need to have it show which text number it is.
    Best regards,
    Kevin R.

  • What is the best way to deploy a java application ??

    Hi,
    I have a question about "deploying" a java application on to a weblogic server.
    Please advise me how to automate in a best way to Deploy 10 different applications(.ear/.war) to 10 different Clusters(containing 40 Managed servers) in wls6.1......and that too in a Production environment.
    I know of using Weblogic Console OR the "deploy" utility to do so ......BUT if i have like 100 clusters .....then i would have to "deploy" 100 times ......
    is there any way that we can BEST AUTOMATE this process.
    Thanks a lot.
    -sangita

    HELLO!!! WELCOME BACK!! I THINK YOU SHOULD USE
    deploy....Hey Thanks.
    is there any network congestion OR any other problem that i can anticipate before i use "deploy" utility. I have heared of some problems(i couldn't remember them now ....because honestly, i couldn't understand them atall when a BEA consultant told me those...).
    so, any problem that may arise ....that i need to think about before deploying ~10 applications to like ~70-80 clusters ....all at a time.
    thanks again for your advise. iam learning to see the big picture of application deployment.
    -sangita

Maybe you are looking for