Problems with Apache JServ

I'm Working with the Apache Jserv module.
When anyone access to any of my servlets, goes to any link, and then return back with the 'Back' button of the browser, the servlet reloads itself, there's no cache or something like that. It means doing the database access again, etc. I'll be greatfull if anybody could help me.
Thanks
Andres Benito

I'm assuming that your database access code in a java bean. If that is the case, you might need to instantiate the bean in your JSP page with SCOPE = "session".
Dennis
null

Similar Messages

  • Right problem with apache and tomcat

    Bonjour;
    I use a user login "apache" to stop/start Apache and tomcat. Because never launching apache et tomcat as root.
    But I have the following problem with apache (file error.log) :
    [Tue May  6 17:26:22 2003] [error] Connection "warpConnection" cannot connect
    [Tue May  6 17:26:22 2003] [error] Cannot open connection "warpConnection"
    [Tue May  6 17:27:01 2003] [error] Re-Trying to deploy connections
    As root the error msg does'nt exists (lost)
    Best regards;
    A+;

    Run it from the shell to see whats wrong:
    # httpd

  • Problems with Apache and custom JSPs

    Hi
    We've made an application on top of IFS, using JWS in our test envirnment. Just before making some stress tests, I'd like to try it using Apache. We're currently having two problems:
    1) I switch to the apache configuration running ifsconfig and not selecting JWS. When I try to access the ifs using http://host/ifs/files, everything goes well except that the "logout" icon doesn't appear. I did a little research and found out that the link goes to /ifs/webui/images/logout.gif. This gives an error in mod_jserv.log, like this one:
    [07/06/2001 22:54:20:315] (ERROR) ajp12: Servlet Error: ClassNotFoundException: webui
    It seems it's trying to find a "webui" class, since in ifs.properties every url that begins with /ifs goes to jserv.
    I don't know if this is a know problem or what should I've check...
    2) This one is more important. We're using some custom JSPs, which we use to edit the properties of some types of documents. Basically, when the user clicks over a file one of our JSP appears. These JSPs call a bean to do some processing, passing the HttpRequest as a parameter. The problem is that when using JWS we get the "path" request variable like in path=/%3A29464
    However, when using Apache we get path=/ifs/files/%3A29464 ( and afterwards we get an exception because the ifsSession.getPublicObject method doesn't work).
    Any hints on this? One way could be to check if the path begins with /ifs/files, but that's not really nice.. and besides I could have the same problem in some other parts.
    It's kind of urgent....
    Thanks
    Ramiro
    null

    Hi,
    The answer to your path problem is that you can make use of API to find out the current path so that it works both with Apache and with JWS. Follow the steps
    1. import the oracle.ifs.adk.http package in your custom jsps
    <%@ page import = "oracle.ifs.adk.http.*" %>
    2. Then within your jsp use the method
    getIfsPathFromJSPRedirect
    <%= oracle.ifs.adk.http.HttpUtils.getIfsPathFromJSPRedirect(request) %>
    This will give you the current path of the object on which you clicked on and which initiates the custom jsp.
    You can look at the CMS application which has made use of this API. URL is
    http://otn.oracle.com/sample_code/products/ifs/sample_code_index.htm
    Choose, sample applicatin -> Content Management system.
    Hope this helps
    Rajesh
    null

  • Problem with Apache reverse proxy after applying SP13 NW

    Hello,
    we have a NW04 EP Portal and a Apache reverse proxy in the DMZ. After applying SP 13 for the portal we get the following error from the reverse proxy:
    Proxy Error
    The proxy server received an invalid response from an upstream server.
    The proxy server could not handle the request GET /irj/.
    Reason: Error reading from remote server
    Apache/2.0.52 (Win32) mod_ssl/2.0.52 OpenSSL/0.9.7e Server at servername.company.de Port 443
    Is is it possible, that there is a problem with sp13?
    Best regards
    Daniel Holstein

    Hi Daniel,
    ok I`ll try to find a solution in parallel and keep you up to date.
    In the following my settings in case I missed something:
    <VirtualHost test.firma.de:443>
    SSLEngine on
    SSLProxyEngine on
    SSLCertificateFile /apache/keys/pac_ssl_qep_dmz_server.crt
    SSLCertificateKeyFile /apache/keys/pac_ssl_qep_dmz_server.key
    ServerName test.firma.de:443
    ServerAdmin [email protected]
    LogLevel debug
    ErrorLog logs/ssl_443_error
    CustomLog logs/ssl_443_access_log common
    ProxyVia Off
    ProxyPreserveHost On
    ReWriteEngine on
    ReWriteLogLevel 0
    ReWriteLog logs//ssl_443_rewrite_http.log
    ProxyPass / https://backend.firma.de:50001/
    ProxyPassReverse / https://backend.firma.de:50001/
    </VirtualHost>
    Regards, Jens

  • Problem with Apache

    I have a problem with starting Apache process.
    There is a log:
    fopen: Invalid argument
    Processing config directory: c:/oracle/product/10.2.0/db_2/apache/modplsql/conf"
    Apache.exe: could not open config directory c:/oracle/product/10.2.0/db_2/apache/modplsql/conf"

    Kostya,
    Is that directory there (does it exist)? And is it readable by the Apache service user?
    Earl

  • Problems with apache

    I downloaded XAMPP Moodle today to do some development and found that apache would not start. It complained that Web Sharing was already turned on in System Preferences and that I should turn it off. When I checked in System Preferences, all indications were that Web Sharing was not enabled. Furthermore, when I tried to turn it on from System Preferences, it would not do so. I opened up http://localhost and my own user account (i.e. localhost/~micah) functions properly and httpd shows up as three processes in Activity Monitor (one under root and two under _www).
    I am confident that I can kill it (using sudo), but I'm not confident that it will not come back in the same way when I restart. Also, how does System Preferences deterine whether or not httpd is running? Does it check the process list? Does it check the port? Is there a pid file? Why does it not show up properly in System Preferences?
    How do I properly fix this situation?

    AstroNauts wrote:
    Thanks for your response!
    Okay, done. Moved it from /Library/WebServer/Documents to /Sites/, stopped working. Now I suppose I have to look into the CGI?
    (apologies, haven't been near this project for a while).
    Okay, it now works!
    Following your hint about the CGI, I uncommented the line below. I don't why it was commented (or should be), but it now works. Had some other help stating it was indeed a problem with how to handel php.
    We're cool, now; thank you for your responses and help!
    Best,
    -Pim
    +LoadModule fastcgi_module libexec/apache2/mod_fastcgi.so+

  • New MS patch problems with Apache & SSL on NetWare web serve

    This week our users began to experience problems accessing our Intranet based on a NetWare 6.5 box running Apache and using a Verisign SSL certificate. IE would say it was connecting but you would never get there. Firefox could get to the server but I had similar problems with Safari & Chrome.
    I discovered that the MS security patch KB980436 had modified some portion of IE and was causing the problem. Uninstalling that patch put us back to where we could once again access our Intranet.
    This seems to have something to do with how IE accesses using the SSL 443 port. When, internally, I accessed this same site using http instead of https it worked perfectly.
    This patch was applied to Win XP and Win 7 with the same bad results.
    If there is a more appropriate place for this post please move it there.

    Have you tried the registry settings to go back to compatible mode? The
    settings are listed in the MS KB article.
    "boomchuck" <[email protected]> wrote in message
    news:[email protected]..
    >
    > This week our users began to experience problems accessing our Intranet
    > based on a NetWare 6.5 box running Apache and using a Verisign SSL
    > certificate. IE would say it was connecting but you would never get
    > there. Firefox could get to the server but I had similar problems with
    > Safari & Chrome.
    >
    > I discovered that the MS security patch KB980436 had modified some
    > portion of IE and was causing the problem. Uninstalling that patch put
    > us back to where we could once again access our Intranet.
    >
    > This seems to have something to do with how IE accesses using the SSL
    > 443 port. When, internally, I accessed this same site using http
    > instead of https it worked perfectly.
    >
    > This patch was applied to Win XP and Win 7 with the same bad results.
    >
    > If there is a more appropriate place for this post please move it
    > there.
    >
    >
    > --
    > boomchuck
    > ------------------------------------------------------------------------
    > boomchuck's Profile: http://forums.novell.com/member.php?userid=28653
    > View this thread: http://forums.novell.com/showthread.php?t=418742
    >

  • CVE-2007-1862 vulnerability - problem with Apache mem_cache.

    I have a problem with CVE-2007-1862, which is an Apache mem_cache vulnerability. Even though my Snow Leopard machine is currently running Apache 2.2.17 and the problem supposedly went away with a much earlier version, the problem has not gone away. By the way, this machine also has Java Update 24 on it, while a similarly configured Snow Leopard machine does not have it (and the Filemaker Pro Admin Console still works on that one).
    Anyway, I am hoping that installing Apache 2.2.18 in a Software update will take care of this problem.
    Does anyone know when Apache 2.2.18 will be in a Software Update?

    Your parameter db_recovery_file_dest look strange
    db_recovery_file_dest = I:\oracle\product\10.2.0/flash_recove db_recovery_file_dest_size= 10737418240
    Did you put a slash (/) instead of a backslash (\) ?
    You're saying "can´t enter this the SYS user"
    It could be your log_archive_dest that is full.
    Do you get an error message?

  • Problem with Apache Plugin

              Hi,
              Hi I am getting some Errors when i am trying to Run the Apache Server. Actual Problem is
              I want to create an Cluster Environment using 2 weblogic Cluster. i read that documentation, using apache-weblogic plug-in, we can improve the cluster performence.
              I follow the Same rules and i install ed the Apache server and i copy the *.so file that required for apache(if i commented out the LoadModule weblogic_module libexec/mod_wl.so , serverrunning fine)
              But whenever i am trying to run apache Server with that
              Its given a error like
              "BUG IN DYNAMIC LINKER ld.so:dynamic-link.h:57: elf_get_dynamic_info: Assertion '! "bad dynamic tag" failed!
              ./apachectl start: httpd could not be started
              If u will give some help about this its most helpful for me.
              ThankQ
              Krishna Botla
              

    Which version of Apache are you using? Are you on Solaris?
              Have you enabled the bootstap module mod_so for apache? Do httpd -l to ensure that.
              Did you use apxs to deploy mod_wl.so?
              Please go through the steps described in the docs again to see if you missed something.
              --Vinod.
              KrishnaKumar wrote:
              > Hi,
              > Hi I am getting some Errors when i am trying to Run the Apache Server. Actual Problem is
              > I want to create an Cluster Environment using 2 weblogic Cluster. i read that documentation, using apache-weblogic plug-in, we can improve the cluster performence.
              > I follow the Same rules and i install ed the Apache server and i copy the *.so file that required for apache(if i commented out the LoadModule weblogic_module libexec/mod_wl.so , serverrunning fine)
              > But whenever i am trying to run apache Server with that
              > Its given a error like
              > "BUG IN DYNAMIC LINKER ld.so:dynamic-link.h:57: elf_get_dynamic_info: Assertion '! "bad dynamic tag" failed!
              > /apachectl start: httpd could not be started
              >
              > If u will give some help about this its most helpful for me.
              >
              > ThankQ
              > Krishna Botla
              

  • Urgent - problems with apache plug-in  proxy resolution

    Hi
    I have apache ver as a plug-in to weblogic clustered servers running WL61 SP1. On
    one of our boxes, we have a tunnelling to point to apache at port 80.
    When I hit from outside (not inside the company extranet), I type in http://www.proxyservername.com/...
    This is routed to apache and apache routes it to weblogic running on the internal
    server eg. johnson.xxx.xxx.xxx.com.
    The problem I am experiencing is that when I send the request -
    http://www.proxyservername.com/../... it gives back response with some session data.
    The next request I sent in is a response.sendRedirect(relative url to a jsp page)
    , this time it drops the old session and points to johnson.xxx.xxx.com:80.
    How can I resolve this client sendRedirect to not flip the server name.
    Thanks in advance.

    What exactly is the redirect to (what is the URL)? Secondly, I'm wondering
    which step in the chain is adding the :80 to the redirect. Is it Weblogic or
    one of the tunnelling doohickies or the browser or what? I would suggest
    trying a non-relative URL (fully qualified) to see if that gets messed with
    too.
    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/
    "arpi" <[email protected]> wrote in message
    news:3ce0500a$[email protected]..
    >
    Hi
    I have apache ver as a plug-in to weblogic clustered servers running WL61SP1. On
    one of our boxes, we have a tunnelling to point to apache at port 80.
    When I hit from outside (not inside the company extranet), I type inhttp://www.proxyservername.com/...
    This is routed to apache and apache routes it to weblogic running on theinternal
    server eg. johnson.xxx.xxx.xxx.com.
    The problem I am experiencing is that when I send the request -
    http://www.proxyservername.com/../... it gives back response with some
    session data.
    The next request I sent in is a response.sendRedirect(relative url to ajsp page)
    , this time it drops the old session and points to johnson.xxx.xxx.com:80.
    How can I resolve this client sendRedirect to not flip the server name.
    Thanks in advance.

  • Problems with Apache Commons Validator

    Hello there,
    I got a quiet big problem using the a.m. validation framework. When defining a validation rule for a property like this:
    <formset>
    <form name="testForm">
    <field property="firstName" depends="isEmail">
    <arg0 key="testForm.firstname"/>
    </field>
    </form>
    </formset>
    and instanciating my validator as follows:
    Validator validator = new Validator(resources, "testForm");
    validator.setParameter(Validator.BEAN_PARAM, name);
    everything works fine! I get a correct ValidatorResult object with a correct validation value.
    But as soons as I define a second Validation on the same property like this:
    <formset>
    <form name="testForm">
    <field property="firstName" depends="isEmail,isBlankOrNull">
    <arg0 key="testForm.firstname"/>
    </field>
    </form>
    </formset>
    Nothing works anymore. When iterating over the ValidatorResults object I only get one ValidatorResult object for "isEmail" with a wrong validation value. There should be two working ValidatorResult objects for "isEmail" and "isBlankOrNull"?!?!
    Are two validation method in the depends-Attribute not allowed (although it is defined in the standard)?
    You can assume, that the used validator class is conform to the given org.apache.commons.TestValidator!
    For any help, thx in advance!
    Greets
    CN

    Hey Aaron,
    Could you post your code for your <field>...</field>?
    Here is an example from: http://www.jsn-server3.com/struts-documentation/userGuide/dev_validator.html
    <field
        property="firstName" indexedListProperty="dependents"
        depends="requiredif">
      <arg0 key="dependentlistForm.firstName.label"/>
      <var>
        <var-name>field[0]</var-name>
        <var-value>lastName</var-value>
      </var>
      <var>
        <var-name>field-indexed[0]</var-name>
        <var-value>true</var-value>
      </var>
      <var>
        <var-name>field-test[0]</var-name>
        <var-value>NOTNULL</var-value>
      </var>
    </field>
    <field
        property="dob"
        indexedListProperty="dependents"
        depends="requiredif,date">
      <arg0 key="dependentlistForm.dob.label"/>
      <var>
        <var-name>field[0]</var-name>
        <var-value>lastName</var-value>
      </var>
      <var>
        <var-name>field-indexed[0]</var-name>
        <var-value>true</var-value>
      </var>
      <var>
        <var-name>field-test[0]</var-name>
        <var-value>NOTNULL</var-value>
      </var>
    </field>
    <field
        property="coverageType"
        indexedListProperty="dependents"
        depends="requiredif">
      <arg0 key="dependentlistForm.coverageType.label"/>
      <var>
        <var-name>field[0]</var-name>
        <var-value>lastName</var-value>
      </var>
      <var>
        <var-name>field-indexed[0]</var-name>
        <var-value>true</var-value>
      </var>
      <var>
        <var-name>field-test[0]</var-name>
        <var-value>NOTNULL</var-value>
      </var>
      <var>
        <var-name>field[1]</var-name>
        <var-value>insureDependents</var-value>
      </var>
      <var>
        <var-name>field-test[1]</var-name>
        <var-value>EQUAL</var-value>
      </var>
      <var>
        <var-name>field-value[1]</var-name>
        <var-value>true</var-value>
      </var>
      <var>
        <var-name>field-join</var-name>
        <var-value>AND</var-value>
      </var>
    </field>
    </form>" Which is read as follows: The firstName field is only required if the lastName field is non-null. Since field-indexed is true, it means that lastName must be a property of the same indexed field as firstName. Same thing for dob, except that we validate for date if not blank.
    The coverageType is only required if the lastName for the same indexed bean is not null, and also if the non-indexed field insureDependents is true.
    You can have an arbitrary number of fields by using the [n] syntax, the only restriction is that they must all be AND or OR, you can't mix. "
    Hope that helps,
    Nate

  • Desperately seeking help. mod_jk problem with Apache 2.0.49 and Tomcat 4.1.

    Hi All,
    I am setting up the connection between Tomcat 4.1 and Apache 2.0.49 since past 3 days without success.
    I downloaded the
    JK 1.2
    JK 1.2 Binary Releases
    mod_jk_1.2.5_2.0.47.dll from apache site.(renamed it to mod_jk.dll as instructed). Now the problem is that after making the other settings in server.xml, httpd.conf,workers.properties
    My Tomcat does not show any error when i uncomment the following code
    <Connector className="org.apache.coyote.tomcat4.CoyoteConnector"
                   port="8009" minProcessors="5" maxProcessors="75"
                   enableLookups="true" redirectPort="8443"
                   acceptCount="10" debug="0" connectionTimeout="20000"
                   useURIValidationHack="false"
                   protocolHandlerClassName="org.apache.jk.server.JkCoyoteHandler"/>
    which is the code for mod_jk2. I don't understand this as I have download the file for JK1.2.
    I am getting the mod_jk.conf file generated but it is generated with the following code.
    <IfModule !mod_jk.c>
      LoadModule jk_module C:/Program Files/Apache Group/Apache2/modules/mod_jk.dll
    </IfModule>
    JkWorkersFile "C:/Program Files/Apache Group/Tomcat 4.1/conf/jk/workers.properties"
    JkLogFile "C:/Program Files/Apache Group/Tomcat 4.1/logs/mod_jk.log"
    JkLogLevel info
    <VirtualHost localhost>
        ServerName localhost
        JkMount /admin ajp13
        JkMount /admin/* ajp13
        JkMount /webdav ajp13
        JkMount /webdav/* ajp13
        JkMount /examples ajp13
        JkMount /examples/* ajp13
        JkMount /tomcat-docs ajp13
        JkMount /tomcat-docs/* ajp13
        JkMount /manager ajp13
        JkMount /manager/* ajp13
    </VirtualHost>
    AGAIN , here I fail to understand why it is generating the jk_module with forward slash.
    (I HAVE GIVEN THE ps=/ property in worker.properties)
    Once I add the path in quotes i.e.
    LoadModule jk_module "C:/Program Files/Apache Group/Apache2/modules/mod_jk.dll"
    I am able to start the apache server. but am failing in the connection test ie. not able to view the examples directory.
    Plz. advise me , I am really going crazy over this stuff.
    Thanks in advance,
    Vishy.

    I am confused, you need to pick modjk2 or modjk.
    I would not use a precompiled modjk2.dll , I have and had security problems.
    I would compile it from source. The instructions are in the source package.
    If you use modjk, your on the right steps, modjk is more difficult to setup. than modjk2.
    For modjk2 all you need is
    LoadModule jk2_module C:/Program Files/Apache Group/Apache2/modules/mod_jk2.dll
    and the workers2.properties in /conf directory, that's it.
    Tomcat 4 and above come 'stock' with ajp workers turned on port 8009.
    First make sure modjk is loaded in Apache.
    Make sure all 30 lines of error page is uncommented on http.conf
    You can do this by http://servername/SOMEbogusAddress
    just go to apache page, then try and load a non-existing page
    You should see at bottom of page:
    Apache/2.0.49 (Unix) mod_jk2/2.0.4

  • Problems with Apache PHP Mysql

    I tried doing an upgrade from Tiger to Leopard server but was having lots of issues while working with the virtual sites I have hosted. I tried a clean install and I was still unable to get Apache php and mysql to work together. Each service was starting fine but when I would try to run a php file it would only display a white page. Under the upgrade instead of displaying the page it was downloading any page I tried to load.
    In the new clean install I enabled phplib5 through server admin but even a simple info.php file with <?php phpinfo(); ?> was only giving me a white page. I dont know if mysql was interfacing with php and apache at this point cause I could not get any further. I mainly just want to be able to run wordpress which is a php blog that uses a mysql database.
    I have heard some people talking about mamp installs and not using the stock web services but I did not know if that would be a smart route to take.

    well the problem come in that you could end out with multiple releases of the same application when the idea is to make one package "stable"
    would you rather have a debain package data base that has seventeen apache's or one stable full featured apache?
    as for a server/stable tree for production machines this is coming. implimentation is not the trouble it is time that is a factor.

  • Problems with Apache Proxy to 3 clustered domains....

    I have 3 different domains (qa, stage, production) running on weblogic 81. I am trying to get Apache to route to the correct domains.
    ie...
    http://192.XXX.XXX.XXX/qa/application connects to 192.XXX.XXX.XX2:7001/application
    http://192.XXX.XXX.XXX/stage/application connects to 192.XXX.XXX.XX2:7002/application
    http://192.XXX.XXX.XXX/application (production) connects to 192.XXX.XXX.XX2:7003/application
    I got that to work, when the URL is first entered into the browser, but the problem is when I try to move around in the qa and stage apps, i lose the "/qa" or "/stage". So it is always goes to the production app.
    httpd.conf -->
    <Location /qa/application>
    SetHandler weblogic-handler
    WebLogicCluster one:7001,two:7001
    PathTrim /qa
    </Location>
    I understand what is happening, when I trim the /qa it is then no longer present in the URL when passed back to the client. So the next movement in the application is sent to
    http://192.XXX.XXX.XXX/application (the production URL). I guess I am unsure how to fix this....any help????
    Thanks...
    Nick

    Hello,
    From this it looks like you may be able to use PathPrepend to jimmy your URLs:
    PathPrepend : String that the plug-in prepends to the beginning of the original URL, after PathTrim is trimmed and before the request is forwarded to WebLogic Server.
    I would suggest it may be a good idea to have a seperate webserver for your production domain. In most places I have worked in every platform is configuered with its own resources e.g staging server has its own webserver and db. This affords the people that use those envrionments greater isolation.

  • Problem with apache plug-in for  WLS 7.0

    I have installed the plug-in follow the "Installing an Configurating the Apache
    HTTP Server Plug-in" that I have found in bea.
    The problem is in the third step of "Installing as a Dinamic Shared Object". The
    Apache 2.0.39 hasn´t the program apxs, and I have added the lines in the httpd.conf
    and copy mod_wl.so in libexec folder. When I test apache server give a problem:
    Invalid Commmand 'AddModule'

    Hi SivaKumar,
    Yes I am using SP13 and I went through the note you referred.
    I am still confused, because my problem is not exactly reflecting in the note.
    I have no applications assgined to my DC as such.
    I am just trying to test this - that if I change the language parameter and set the condition which is false, that change should not reflect on the client.
    What do you think I can do about it?
    Ankur

Maybe you are looking for