Proxying multiple virtual websites from IIS to multiple WLS

I have an IIS 5.0 server with four virtual sites, each with it's own IP address. I have two seperate WLSs (one is 5.1 and one is 6.1) that I want the sites to proxy to. I followed the instructions in "Installing and Configuring the Microsoft Internet Information Server (ISAPI) Plug-In" but can only have one virtual server running at a time for it to forward. What am I doing wrong?

I see this information in the admin manual for version 5.1 and 6.1 but not for version 6.0. Was it just omitted by accident or is there a different process for handling multiple IIS virtual web sites in version 6.0?

Similar Messages

Multiple weblogic instances using IIS to proxy the requests

I have a need to run multiple weblogic instances using IIS as the web
server.
Does anyone know if this is possible. I am aware and have been able to run
multiple weblogic servers as an nt service,
but in those cases weblogic was being called and IIS was being bypassed.
Now the user base wishes to use IIS to proxy the requests.
I have been able to set IIS to forward using the .jhtml and .jsp file
extensions to the iisproxy.dll, but have not been able to forward these
requests to multiple weblogic instances. When the request is made it is
only looking in the first instance of weblogic for the request.
Is it possible to use weblogic and IIS with two or more weblogic instances,
without weblogic acting as a webserver.
I dont believe this is possible but I need to find out if this really is the
case.
Thanks for any help you can provide.
Ed

You need run each IIS web server instance in its own memory space. This can
be set through the Management Console for IIS.
Keith
Mike Kendzierski <[email protected]> wrote in message
news:[email protected]..
We tried to do this & Weblogic doesn't support this config. We opened upa
support call with them regarding this, but there is a problem with the way
that IIS & NT proxies .JSP pages to the weblogic servers. It's very
unstable this way & I wouldn't recommend this config.
We switched to Apache with virtual hosts to do this & it works fine. It's
probably not the answer you want to hear, but it works very well.
If you have any questions, just ask away.
"Ed" <[email protected]> wrote in message
news:3a296c2d$[email protected]..
We are using IIS4 with weblogic 4.5.1 sp11,
I thought I read someplace, their was a problem between wls 4.5.1 sp7
and
wls 5.1 where it didnt look in the folder where the dll was.
I will give it a try on monday.
Peter Mergaerts wrote in message <[email protected]>...
We tried to do the same thing.
We copied iisproxy.dll and iisproxy.ini to server1 and let IIS use this.dll
instance for proxying .jsp etc.
We copied the iisproxy.dll and iisproxy.ini to server 2, renamed
iisproxy.dll to iisproxy2.dll (to make sure another instance was used)
and
let the second iis site use iisproxy2.dll, but that didn't worked. (It
doesn't read the iisproxy.ini file located in the same directory as the.dll
file, altough in the manual it says it checks this first.) Now, I thinkit's
possosible to use several iisproxy.dll's as long as you rename it so
each
get it's own instance, the only problem is that each .dll doesn't readthe
correct iisproxy.ini file. (BEA, is this for SP7?)
Regards,
Peter Mergaerts

Multiple virtual webservers with same files

Hello friends... I need some help to design my enviroment.
I have here, 12 webservers which have independent files. My intent is use the same files for all of them and use shared configuration from iis.
They are all virtual servers runing on hyper-v 2008 r2. How can i use the same files for those servers? I was thinking on network share but I dont know if its the best way to do this.
tried to use a passthrough disk on them but didnt work as i spected... theres any way to one server have write permission and the others just read and when other need to write something, the permission change from one to another?
Im kinda lost here...

"how can i use a unique disk to multiple servers access data as webserver?"
The answer is in the article referenced by Aidan. You set up a file share. The beauty of a file share is that it can also be served out by a cluster, meaning that the data is highly available. If you just try to share a single physical/virtual
disk, if anything happens to that, all web servers go down.
. : | : . : | : . tim

Looking to Set Up Multiple Virtual Desktops

Hi, I'm a newb looking to set up multiple virtual desktops for my employees. What would be the best windows image to use and how do I install Chrome onto the machines?

Hi Shru84;
Welcome to Azure. Azure Virtual Machine as the name suggest are Virtual Server which function as your regular server but does not have a physical presence but is hosted on Azure Platform.
We have a variety of OS flavors, there is nothing which is good or bad, its more of a question of your requirement. Please note that the desktop Version of Windows 7/ 8
are at the moment only available for MSDN subscriber.
As for Chrome Installation its the same as installing it on a regular machine, that is opening up a web browser, loggin into chrome download site and installing it. ( http://www.google.com/chrome/ )
Here is an example:
his tutorial shows you how easy it is to create an Azure virtual machine (VM). This tutorial uses a Windows Server image, but that's only one of the many images available through Azure. This includes Windows operating systems, Linux-based operating systems,
and images with installed applications. The images you can choose from depend on the type of subscription you have. For example, desktop images may be available to MSDN subscribers.
You can also create Windows VMs using
your own images as templates. To learn more about Azure VMs, see
Overview of Azure Virtual Machines.
NOTE:
You need an Azure account to complete this tutorial:
You can
open an Azure account for free: You get credits you can use to try out paid Azure services, and even after they're used up you can keep the account and use free Azure services, such as Websites. Your credit card will never be charged, unless you explicitly
change your settings and ask to be charged.
You can
activate MSDN subscriber benefits: Your MSDN subscription gives you credits every month that you can use for paid Azure services.
How to create the virtual machine
This section shows you how to use the From Gallery option in the Management Portal to create the virtual machine. This option provides more configuration choices than the
Quick Create option. For example, if you want to join a virtual machine to a virtual network, you'll need to use the
From Gallery option.
NOTE:
You can also try the richer, customizable Azure Preview Portal to create a virtual machine, automate the deployment of multi-VM application templates, use enhanced VM monitoring and diagnostics features, and more. The
available VM configuration options in the two portals overlap substantially but aren't identical.
Sign in to the Azure Management Portal. Check out the
Free Trial offer if you don't have a subscription yet.
On the command bar at the bottom of the window, click New.
Under Compute, click Virtual Machine, and then click
From Gallery.
The first screen lets you Choose an Image for your virtual machine from one of the lists in the Image Gallery. (The available images may differ depending on the subscription you're using.) Click the arrow to continue.
The second screen lets you pick a computer name, size, and administrative user name and password. If you just want to try out Azure Virtual Machines, fill in the fields as shown in the image below. Otherwise, chose the tier and size required to run your
app or workload. Here are some details to help you fill this out:
New User Name refers to the administrative account that you use to manage the server. Create a unique password for this account and make sure to remember it.
You'll need the user name and password to log on to the virtual machine.
A virtual machine's size affects the cost of using it, as well as configuration options such as the number of data disks you can attach. For details, see
Virtual Machine and Cloud Service Sizes for Azure.
The third screen lets you configure resources for networking, storage, and availability. Here are some details to help you fill this out:
The Cloud Service DNS Name is the global DNS name that becomes part of the URI that's used to contact the virtual machine. You'll need to come up with your own cloud service name because it must be unique in Azure. Cloud services are important
for scenarios using
multiple virtual machines.
For Region/Affinity Group/Virtual Network, use a region that's appropriate to your location. You can also choose to specify a virtual network instead.
NOTE:
If you want a virtual machine to use a virtual network, you must specify the virtual network when you create the virtual machine. You can't join the virtual machine to a virtual network after you create the VM. For more information, see
Azure Virtual Network Overview. - For details about configuring endpoints, see
How to Set Up Endpoints to a Virtual Machine.
The fourth configuration screen lets you configure the VM Agent and some of the available extensions. Click the check mark to create the virtual machine.
NOTE:
The VM agent provides the environment for you to install extensions that can help you interact with or manage the virtual machine. For details, see
Using Extensions.
After the virtual machine is created, the Management Portal lists the new virtual machine under
Virtual Machines. The corresponding cloud service and storage account also are created and are listed in those sections. Both the virtual machine and cloud service are started automatically and the Management Portal shows their status as
Running.
Click Virtual Machines, and then select the appropriate virtual machine.
On the command bar, click Connect.
Click Open to use the Remote Desktop Protocol file that's automatically created for the virtual machine.
Click Connect to continue.
Type the credentials for the administrative account on the virtual machine, and then click
OK.
TIP:
In most cases, you'll use the user name and password that was specified when the virtual machine was created. Check the user name to make sure it has the correct domain information:
If the VM belongs to a domain at your organization, make sure the user name includes the name of that domain.
If the VM doesn't belong to a domain, either remove any domain information by starting the line with '\' or use the VM name as the domain name. For example,
\MyUserName or MyTestVM\MyUserName.
If the VM is a domain controller, type the user name and password of a domain administrator account for that domain.
Click Yes to verify the identity of the virtual machine.
You can now work with the virtual machine remotely.
Regards;
Prasant

Oracle Portal Virtual Host Enhacement for Multiple Virtual Hosts

I open the TAR Number: 4326752.996
The TAR is related to " REQUEST ASSISTANCE FOR MULTIPLE VIRTUAL HOSTS AND MULTIPLE PAGE GROUPS CONFIG. "
This is the Scenario: We Have Oracle Portal installed in a single Box. yes I know I know... but Not everybody is Yahoo or Google for Clusters and all the good stuff, we just want our website managed by Oracle Portal since all our ERP is running Oracle Database. Anyway. 1 Box running Oracle Portal Infrastructure and MidTier.
It turns out that our previous configuration we have the LAMP Solution with PHP, Apache with Virtual Servers configuration very trivial for a regular web server however we have about 6 Virtual domains.
For example:
www.abc.com
www.def.com
www.ghi.com
www.jkl.com
www.mno.com
www.pqr.com
We also like to post all our Intranet Information in this great Portal:
http://intranet.com
Where intranet.com is the same server but setup in our Internal DNS without be a Real Internet Register Domain.
So bottom line we start doing great with Install our single box install, later [GRRRRRR does not come by default in port 80 GRRRRRR, ok need to move and setup the virtual host to resolve our main External website www.abc.com ]
For starters ok it worked we have our main website running Oracle Portal.
Now is time to go to the step number 2.
Setup another Page Group in Portal to be the Page_Group_INTERNAL.
I figure if Apache can setup Virtual Hosts pointing to Independent parts of the File System without be shared. I was expecting some kind of custom configuration where I can setup httpd.conf to point to some specific PageGroups to make sure
www.abc.com => Page_Group_External
never share information with:
intranet.com => Page_Gorup_Internal
My TAR was very quick closed with the Argument that is not what It supposed to do.
Since a virtual host is a common thing to do on Linux and Apache, I wonder if You guys could list this as one of the default features to have multiple virtual domains assosiated with Page Groups that they never talk or share Resources I think that would be really great.
In the mean time I guess we have no option but to put a password for every internal page or take the risk of share our Information to the Internet. Perhaps designate another box with Oracle Portal but In that case Is not really cost/bennefit to use a killer technology for our external site since we just have no more than 30 pages.
Anyway just figure drop you guys a note any advise will be really appreciated.
Best Regards Dino.

Hi Martin,
1. To change the target URL of KM resources you can to the following:
Access the URL Generation Service Configuration by navigating to:
System Administration -> System Configuration -> Knowledge Management -> Configuration
Content Management -> Global Services -> URL Generator Service
Change the Host property to "portal.company.com".
2. You change the Port by opening the Visual Administrator and going to : Dispatcher -> Services -> HTTP Provider
Change there the "Ports" entry from (Port:50000,Type:http)to (Port:80,Type:http). Then go on "Save".
3. To remove the "/irj" from the portal URL you can create a index.jsp with the following content, and save it in the J2EE Documentation folder:
<Drive>:\usr\sap\<SAPSID>\JC<J2EE_instance_number>\j2ee\cluster\server0\apps\sap.com\com.sap.engine.docs.examples\servlet_jsp\_default\root
<%
//Send the redirect using server redirection (not client script)
//This will also send the HTTP GET parameters
String queryString = request.getQueryString() ;
if (queryString==null ) {
response.sendRedirect
("irj/servlet/prt/portal/prtroot/com.sap.portal.navigation.portallauncher.default") ;
} else {//redirect with HTTP GET parameters
response.sendRedirect
("irj/servlet/prt/portal/prtroot/com.sap.portal.navigation.portallauncher.default?" + queryString) ;
%>
Hope this helps,
Robert

Setting geo.enabled to false does not stop websites from knowing location. Seems to be related to my proxy setting.....Any ideas on how to completely disable this?

Setting geo.enabled to false does not stop websites from knowing location. It changes based off of what proxy I am using. Is there a way to completely disable this?

I use Comcast in the Chicago area market and have a problem with the "Opt Out" stopping about once a year or so. It appears as being enabled at the Comcast Opt Out page, but the unresolved addresses are still going to the DNS "domain helper" service.
What works for me is to turn off the opt out preference, log off from that Comcast account, and close Firefox. When I restart Firefox, log in to Comcast, and then set the pref to "opt out" again - it works as intended and doesn't cause me a problem until it happens like 10 to 12 months later.
As far as why it seems to affect browsers in a different way, I have no idea. I do have Safari, Opera, and IE 8 installed, in addition to multiple versions of Firefox, but I don't use those other browsers enough to have seen a difference. I am usually opening pages from within Firefox in those other browsers to view the difference, ''or to print something that Firefox screws up in the Print Preview''.

SAP EP calling ASP screen - asking multiple logon prompt from internet

Hi,
We have EP page which calls multiple ASP page from the IIS server. When the users are in internet the system prompt for multiple prompt from the IIS server.
1. May i know to make it as a single logon
2. is it possible to do SSO between IIS and EP 7.3
Thanks

Hi,
you can use the SSO22KerbMap ISAPI Filter to accomplish you requirement. Please refer to
Using SAP Logon Tickets for Single Sign on to Microsoft based web applications
and
Single Sign-On of Windows-based Web Service Clients using SAP Logon Tickets
for more details.
Alterantively you can try with SAML (see Single Sign-On with SAML 2.0 - Security and Identity Management - SCN Wiki for more details), but then you should implement SAML Service provider on the IIS/ASP side, since unless I'm mistaken the IIS not provide it by default. Maby this link can be helpful for your: An Open Source ASP.NET SAML2 Service Provider | Passion for Coding
Best regards,
Artem

[svn] 949: Bug: BLZ-96 - When sending a HttpService request from ActionScript with multiple headers with the same name , it causes a ClassCastException in the server

Revision: 949
Author: [email protected]
Date: 2008-03-27 07:12:59 -0700 (Thu, 27 Mar 2008)
Log Message:
Bug: BLZ-96 - When sending a HttpService request from ActionScript with multiple headers with the same name, it causes a ClassCastException in the server
QA: Yes - try again with legacy-collection true and false.
Doc: No
Checkintests: Pass
Details: Another try in fixing this bug. When legacy-collection is false, Actionscript Array on the client becomes Java Array on the server and my fix yesterday assumed this case. However, when legacy-collection is true, Actionscript Array becomes Java ArrayList on the server. So added code to handle this case.
Ticket Links:
http://bugs.adobe.com/jira/browse/BLZ-96
Modified Paths:
blazeds/branches/3.0.x/modules/proxy/src/java/flex/messaging/services/http/proxy/RequestF ilter.java

Hi all!
Just to post the solution to this if anyone ever runs accross this thread...
For some reason i had it bad the first time, don't have time right now to see why but here is what worked for me:
HashMap primaryFile = new HashMap();
primaryFile.put("fileContent", bFile);
primaryFile.put("fileName", uploadedFile.getFilename());
operationBinding.getParamsMap().put("primaryFile", primaryFile);
HashMap customDocMetadata = new HashMap();
HashMap [] properties = new HashMap[1];
HashMap customMetadataPropertyRoom = new HashMap();
customMetadataPropertyRoom.put("name", "xRoom");
customMetadataPropertyRoom.put("value", "SOME ROOM");
properties[0] = customMetadataPropertyRoom;
customDocMetadata.put("property", properties);
operationBinding.getParamsMap().put("CustomDocMetaData", customDocMetadata);
Basically an unbounded wsdl type is an array of objects (HashMaps), makes sense, i thought i had it like this before, must have messed up somewhere...
Good luck all!

[svn] 931: Bug: BLZ-96 - When sending a HttpService request from ActionScript with multiple headers with the same name , it causes a ClassCastException in the server

Revision: 931
Author: [email protected]
Date: 2008-03-26 11:31:01 -0700 (Wed, 26 Mar 2008)
Log Message:
Bug: BLZ-96 - When sending a HttpService request from ActionScript with multiple headers with the same name, it causes a ClassCastException in the server
QA: Yes - we need automated tests for this basic case.
Doc: No
Checkintests: Pass
Details: RequestFilter was not handling multiple headers with the same name properly.
Ticket Links:
http://bugs.adobe.com/jira/browse/BLZ-96
Modified Paths:
blazeds/branches/3.0.x/modules/proxy/src/java/flex/messaging/services/http/proxy/RequestF ilter.java

Hi all!
Just to post the solution to this if anyone ever runs accross this thread...
For some reason i had it bad the first time, don't have time right now to see why but here is what worked for me:
HashMap primaryFile = new HashMap();
primaryFile.put("fileContent", bFile);
primaryFile.put("fileName", uploadedFile.getFilename());
operationBinding.getParamsMap().put("primaryFile", primaryFile);
HashMap customDocMetadata = new HashMap();
HashMap [] properties = new HashMap[1];
HashMap customMetadataPropertyRoom = new HashMap();
customMetadataPropertyRoom.put("name", "xRoom");
customMetadataPropertyRoom.put("value", "SOME ROOM");
properties[0] = customMetadataPropertyRoom;
customDocMetadata.put("property", properties);
operationBinding.getParamsMap().put("CustomDocMetaData", customDocMetadata);
Basically an unbounded wsdl type is an array of objects (HashMaps), makes sense, i thought i had it like this before, must have messed up somewhere...
Good luck all!

How to call multiple stored procedure from same DB Adapter

Hi,
I want to invoke 3 stored procedures from my message flow in BUS . I created a DB adapter and invoked 1 successfully . Now I don't to configure 2 other DB adapters for calling other 2 stored procedures . I want some how to be able to call the other 2 from the first DB adapter only . Somehow I need the flexibility to have procedure names as the operation names in 1 wsdl and be able to configure this in JDeveloper for this at time of DB adapter configuration . I am using 11G . Is it possible ?

You can probably create a DB adapter with operation type selected as Execute Pure SQL and write SQL code to invoke the stored procedures depending on the procedure name (which you can get as one of the parameters of DB Adapter).
However, Even if you were able to do it, the DB Adapter schemas are generated based on stored procedure parameter list and types. If you use Execute Pure SQL type of DB Adapter the schema will be generated based on the parameters which you are using in your custom SQL code within DB Adapter. Which means that if you want to add a new stored procedure as target which has a different name and different parameters then you will need to regenerate the DB adapter and update your custom SQL code. This also means that you will need to do regression testing interfaces to all stored procedures whenever there is any change in this DBAdapter.
Now, my question is why do you want to invoke multiple stored procedures from same DB adapter?
This is not a good way to design integration solutions and makes your services resistance to change as it will take more time to change and test.
If all that you want is to have a web service which can have different operations for different stored procedures then you should create three business services and create a wrapper proxy service which has one operation for each stored procedure, within the proxy transform and call the correct backend service (you can use operational branch). This way you have a modular code where each module (read interface to one stored procedure) can be independently modified and tested.

Tutorial Announcement .:: How To Create a Multiple Style Website ::.

Hello Everyone...
Building a multiple style website is amazing thing, where you can have a new look for your website every day, or even everytime you want your website fully changed from style to another, and give the ability to your website visitors to pick one of available styles they comfortable with.
Today herein this tutorial we will learn how to build a dynamically switched style website application, and i mean dynamically here by storing a number of styles into the database table by uploading more than one [.css] files to your server throught File manipulation process, then define which one of these styles is the default one, the default style automatically checked when every page loaded in the browser by checking an external file contains the default style information set by the site administrator.
After complete this tutorial you will be able to:
1- Uploading new style to your website by using easy Online Upload script.
2- Updating curent style files.
3- Choose which style is the default style and which is not.
4- Allow visitors to select one of available styles to switch the whole site look to it.
Note: you cannot define more than one default style, otherwise you will get un stylized page at all.
:: Go To Online Demo ::
:: Go To Tutorials ::
Best Regards
Waleed Barakat
Developer-Online Creator and programmer
http://www.developer-online.com

Hey Waleed,
thanks again for this great stuff -- and thanks for removing the login protection :-)
Cheers,
Günter Schenk
Adobe Community Expert, Dreamweaver

WAS and multiple virtual Host

I am currently evaluating Oracle WAS on Linux.
It works fine except if I try to have Spyglass serve more than
one virtual host. When I do this it stops with an Internal error
after serving a few files.
Q. Does anyone have Spyglass serving for more than one virtual
host on the same port.
Q. What could be the problem
I use RH 5.1
Johnny Verhoeven
null

Web is stupidly easy - Apache has extensive support for multiple virtual hosts in a single server and will happily support as many domains/hostnames as you can throw at it.
Mail is a little trickier since it needs to hook into some kind of directory server to validate user addresses.
Mac OS X's Open Directory does support this, but it's a little cumbersome (basically every user has to exist in the main Open Directory domain as well as having a secondary email address(es) in the other domain(s).
This can get especially tricky when you have users with the same shortname in different domains - for example if you give [email protected] a secondary email address of [email protected] and you then get a new user [email protected], you can't give the third 'joe' the address 'joe' in Open Directory since that's already taken. This means he'll have to login using different credentials from his email address. Confused yet? You bet.
If you're prepared to eschew Open Directory you can just use Postfix's built-in support for virtual hosting, but it's a little more complex to setup and support and is entirely command-line based (i.e. there's no GUI for setting it up or supporting it).
So, for that reason you might be prepared to setup multiple virtual machines for each domain. It really depends on how much the domains are related (if at all).

Multiple virtual host impacting performance

Is there a performance issue with relation to the number of virtual hosts
which one server is hosting? If there is, how many can an average server
handle before one would see a performance hit?
Thank you.

Web is stupidly easy - Apache has extensive support for multiple virtual hosts in a single server and will happily support as many domains/hostnames as you can throw at it.
Mail is a little trickier since it needs to hook into some kind of directory server to validate user addresses.
Mac OS X's Open Directory does support this, but it's a little cumbersome (basically every user has to exist in the main Open Directory domain as well as having a secondary email address(es) in the other domain(s).
This can get especially tricky when you have users with the same shortname in different domains - for example if you give [email protected] a secondary email address of [email protected] and you then get a new user [email protected], you can't give the third 'joe' the address 'joe' in Open Directory since that's already taken. This means he'll have to login using different credentials from his email address. Confused yet? You bet.
If you're prepared to eschew Open Directory you can just use Postfix's built-in support for virtual hosting, but it's a little more complex to setup and support and is entirely command-line based (i.e. there's no GUI for setting it up or supporting it).
So, for that reason you might be prepared to setup multiple virtual machines for each domain. It really depends on how much the domains are related (if at all).

Random question: Why are there multiple virtual consoles?

I'm very interested in every single detail of some things and that has spawned a random question.
Why do we have multiple virtual consoles(proper term?) ? i.e. if I am looking at my desktop I can simply hit ctrl-alt-f3 for example, and poof! I'm at a new login prompt.
What is/are the purpose(s) for this feature? Can someone name off some scenarios?
Thanks!

Forgotten Path wrote:
How so? I tried running
virtualbox --startvm Windoze
on a virtual console and received an error about finding the X server.
Here is my ~/.xinitrc
#xset b off
setxkbmap -option ctrl:nocaps
case $WM in
kde)
ck-launch-session dbus-launch startkde
gnome)
ck-launch-session dbus-launch gnome-session
compiz)
~/bin/compiz&
ck-launch-session dbus-launch compiz ccp
#ck-launch-session dbus-launch fusion-icon
lxde)
~/bin/lxde&
ck-launch-session dbus-launch startlxde
openbox)
~/bin/openbox&
ck-launch-session dbus-launch openbox
e17)
ck-launch-session dbus-launch
enlightenment_start
vb)
VirtualBox -startvm "Windows XP" -fullscreen
xfce4)
ck-launch-session dbus-launch startxfce4
i3)
ck-launch-session dbus-launch i3
ck-launch-session dbus-launch startxfce4
esac
From the console, use :
WM=vb startx
or, if there is already something running on Screen 0 (OpenBox, Xfce4, whatever) simply:
WM=vb startx -- :1
Note that the examples start VitutalBox. I could just as easily set WM to any of the other environments defined in my .xinitrc.
I have also made this work straight from kdm. I think I created a custom session, but I may have created a .desktop file in /usr/share/apps/kdm/sessions/
I am not using kdm on this machine and I don't remember
edit: Typo
Last edited by ewaller (2012-02-11 23:14:09)

Multiple CSV exports from the one button or pl/sql procedure?

I need to have multiple csv exports from the one press of a button. The easiest way I found to do this is it to use javascript to popup three windows, each as a CSV link. This is a bit ugly though, and leaves the browser popup windows open when the file has been downloaded.
I guess I could also make a solution based on branching, but I think that would be difficult to maintain and reeks of bad design (im not a fan of this spagetti GOTO style code!).
I implemented Scott's custom CSV as found here: http://spendolini.blogspot.com/2006/04/custom-export-to-csv.html
However I would like to know if its possible to download more than one file using this method. I could not work out how to do this .
Has anyone got any ideas? Simply repeating the code puts the second table into the original csv file. Is there a way to 'reset' the htp writer or smoething?
Any help greatly appreciated,
Alex

Sorry for the confusion - I guess I mean its easy in .NET because you can simply compress files together and then send 1 zip file down as the response. See http://www.developer.com/net/net/article.php/11087_3510026_2 for details.
I guess I could ask how to do this in APEX - but it seems to me that my original wording addresses the concept at a much more abstract level. I may not find the best solution for my problem if I just asked 'how can I dynamically zip together three tables as seperate files and send them to the client?'. I also suspect that this method is not possible in APEX without custom packages. Please prove me wrong!
I guess even if I could find some kind of javascript that didnt open a new window, but was a direct download to the CSV, that would be a good compromise. At the moment when you click on the link, three windows come up and stay blank until the files are ready for downloading. Then after the files have been downloaded the windows must be shut manually. Yes, I could use javascript to make the windows 1x1 pixel perhaps, and then shut them after a predetermined timeframe - but this is hardly an elegant solution!
Thanks for your responses.

Proxying multiple virtual websites from IIS to multiple WLS

Similar Messages

Maybe you are looking for