Physical Directory in WE21

Hi,
I had created an XML File port and provided c:/test folder as the phyical directory.
Although when I had executed the IDoc, the status records says that it has writted the file , i could not see that in c:/test folder.
I tried logical directory by providing /tmp path in al11. But, no luck.
Is there anything I would need to do with physical or logical directory assignment.
I had been following steve's blog Collecting IDocs without using BPM

hi Krish,
a better way to transfer collected Idocs from ECC to PI, is to use Steve blog + port XML-HTTP. See this doc:
[How Tou2026 Send Multiple IDocs within one XI message.pdf|https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/30ea2fdf-f047-2a10-d3a2-955a634bde6b]
Else about your physical directory, remind that you can only defined a directory of your PI server (not your desktop of course).
So see with admin team to get the good one, because maybe your directory is not yet mount with AL11 tcode.
Note: by using port XML-file or XML-HTTP file, you loose the possibility to return easily a technical Acknowledgment to ECC (idoc ALEAUD), as it's no more a real idoc (as you don't use an idoc port in ECC), except if you create the corresponding interface.
Regards;
Mickael
Edited by: Mickael Huchet on Oct 17, 2011 9:49 AM

Similar Messages

How to upload file to server physical directory

Dear Sir,
My Software Details:
Oracle9i Release 1
Oracle9iAS 1.0.2.2.2
Solaris 8
I have a web application developed using PL/SQL.
How do i upload a file from user to a server physical directory through Oracle9iAS without storing in database?
I can do this if file is uploaded to a 9ias upload table.
But i'm looking for solution on uploading file straight to the physical server directory.
Please advise.
Thanks.
Regards,
Jap Boon Churn

Hi Jap,
You won't be able to do this with modplsql if that is what you are asking. modplsql only uploads file into the database.
With that said, however, you can write your own CGI/Servlet/Perl program that does the uploading of files into the server physical directory. In your program, you will need to handle HTTP POST requests so that you can parse out the file contents and write it out onto the file system.
I am sure there is already some CGI/Servlet/Perl program out there that does this. If you search on Google, you should be able to find some and modify it to your needs. Hope that helps. Thanks.
Eric.

Mapping of Web App context root and the physical directory of the web app

I'm running Weblogic 7.0 on Windows2000.The physical directory of my web application
is D:\WL8\weblogic81\TestDeploy\build\TestWebApp and under these directory I have
my JSPS, static HTML and WEB-INF. I define the context path of this web app in
the weblogic.xml ;-
<weblogic-web-app>
<context-root>/testapp</context-root>
</weblogic-web-app>
As a result of deploying this web app in the server (or it may be created manually
also), the following entry gets inserted in the server's config.xml ,-
<Application Deployed="true" Name="TestWebApp"
Path="D:\WL8\weblogic81\TestDeploy\build" TwoPhase="true">
<WebAppComponent Name="TestWebApp" Targets="myserver" URI="TestWebApp"/>
</Application>
Now, whenever I make a request of the form "http://localhost:7001/testapp/..",
it's properly executing my web app. My question is, how does the container knows
that for any request for the web app with context path as 'testapp', it has to
server files from D:\WL8\weblogic81\TestDeploy\build\TestWebApp. In the above
process, nowhere such mapping is specified. I expected something like Tomcat's
server.xml, where in docbase we clearly specify this mapping between the context
path and the physical directory. Please help.

Let me give some more details and hopefully this will make things clearer.
Say you deploy /foo/bar/myweb.war and in myweb.war you configure a
context-root of /rob
During deployment, the server creates an ApplicationMBean with a path of
/foo/bar/. It then creates a WebAppComponent with a uri of myweb.war.
Next, deployment calls back on the web container and tells it to deploy
the WebAppComponent. The web container reads the myweb.war, parses
descriptors etc. The web container then updates its data structures to
register that myweb.war has a context path of /rob. (It has to figure
out all the other servlet mappings as well.)
When a request for /rob/foo comes in, the web container consults its
data structures to determine which webapp and servlet receives the
request. This is not a linear search of all webapps and servlets.
There's much better ways to do pattern matching.
Hope this clears things up. Let me know if you still have questions.
-- Rob
Arindam Chandra wrote:
Thanks for the answer. Still one thing is not clear. Whatever context path I declare
for my web app as the value of <context-root> element in the weblogic.xml (in
my example it's "/testapp"), it is no where mapped with the "URI" attribute (or
any other attribute, sub-element whatsoever in the <Application> element).
Application Deployed="true" Name="TestWebApp"
Path="D:\WL8\weblogic81\TestDeploy\build" TwoPhase="true">
<WebAppComponent Name="TestWebApp" Targets="myserver" URI="TestWebApp"/>
</Application>
So when a request of the form http://myweblogic.com:7001/testapp/... arrives at
the server, how does the server knows that it has to serve this request with files
from D:\WL8\weblogic81\TestDeploy\build\TestWebApp ? It should not be like the
web container iterates thru all the web application entries in config.xml and
tries to match with one context-root declaration. I repeat, I expected some mapping
similar to Tomcat's server.xml, where in the <docbase> element u clearly specify
the mapping between the context path and the physical directory
Rob Woollen <[email protected]> wrote:
Arindam Chandra wrote:
I'm running Weblogic 7.0 on Windows2000.The physical directory of myweb application
is D:\WL8\weblogic81\TestDeploy\build\TestWebApp and under these directoryI have
my JSPS, static HTML and WEB-INF. I define the context path of thisweb app in
the weblogic.xml ;-
<weblogic-web-app>
<context-root>/testapp</context-root>
</weblogic-web-app>
As a result of deploying this web app in the server (or it may be createdmanually
also), the following entry gets inserted in the server's config.xml,-
<>So the server will look for your web application at the Application Path
(D:\WL8\weblogic81\TestDeploy\build|) + the web uri (TestWebApp). So
it
maps the context-root you've specified /testapp to that path.
It's a little clearer in the case where you had a full-fledged EAR.
Then you'r application path would map to the "root" of the EAR, and the
uris would point to the various modules (eg webapps.)
-- Rob
Now, whenever I make a request of the form "http://localhost:7001/testapp/..",
it's properly executing my web app. My question is, how does the containerknows
that for any request for the web app with context path as 'testapp',it has to
server files from D:\WL8\weblogic81\TestDeploy\build\TestWebApp. Inthe above
process, nowhere such mapping is specified. I expected something likeTomcat's
server.xml, where in docbase we clearly specify this mapping betweenthe context
path and the physical directory. Please help.

Creating Physical Directory

hii...
I am using CREATE DIRECTORY method.
CREATE DIRECTORY MYDIR AS 'C:\TEMP1';
it is working. But, i want to create physical directory if TEMP1 folder isn't in the C: drive.
How can i do this? do you have any idea? thanx

Oracle implemented the DIRECTORY model primarily for security reasons, so normal users should not be issuing CREATE DIRECTORY statements. A privileged user should create the directory and then give normal users appropriate permissions on the directory. The privileged user should know whether or not the physical directory exists.
Just in case you did not know, the O/S directory pointed to by an Oracle DIRECTORY object needs to be on the same server as the Oracle instance itself.
John

Physical directory and logical directory

What is the difference between physical directory and logical directory in BPEL

Hi,
Physical : We need to Specify the complete directory.
Example: <Drive name>:\Temp\data
Logical: We need to specify a logical name for directory. It is defined as a property in the 'bpel.xml' file.
Example: if you specify 'READ' as the logical name, the following property is added to the bpel.xml file: <property name="READ"></property>
After this, u must have define the directory path.
For example, <Drive name>:\Temp\data for the file adapter: <property name="MyDirectory"><Drive name>:\Temp\data </property>
cheers,
Abhi

How will i create physical directory in oracle pl/sql using java

how will i create physical directory in oracle pl/sql using java

This is, I suspect, related to Re: how will i overcome from this error java.io.IOException. Quite why you felt the need to have two partial threads instead of one thread with a complete set of information is beyond me. Perhaps this muddled thinking is indicative....
Anyway, if I was looking to do this particular task in Java I would look to using the specific command File.mkdir() rather than trying to use a generic exec approach.
Cheers, APC

Directory logical path to directory physical path

Hello experts,
I have question related to getting path from defined earlier logcal path in FILE transaction.
I know there are FMs called 'FILE_GET_NAME' and 'FILE_GET_NAME_AND_LOGICAL_PATH' but they are not solving my problem.
Let's say I know only physical file name and logical file directory, but I don't know the full physical path of file.
I now want to build a program where user writes down the physical name of the file and in the code side there is defined logical file directory from which we can get the physical directory path and concatenate it with entered filename. Is there a possibility to do that?
Thank you in advance.
Best regards,
Andrew

function module FILE_GET_NAME_USING_PATH is doing exactly that.

Creation of rfilr ports in WE21 from LSMW IDOC Processing

Hi,
I am trying to defile a file port for INBOUND IDOC processing status in LSMW.
Now when I go to create a file port in WE21, I give some dummy Physical directory name. But what should I give in
Function module field and Outbound file field ? Is it compulsary to put something in these fields...
Please help.....

Hi,
Function module field is used to define the path functionality of file generated.
EDI_PATH_CREATE_CLIENT_DOCNUM is the recommned fucntion by SAP .
Function is used to prevent unprocessed files being overwritten.Tihs function generates file directory path dynamically.
Hope this helps.
Thanks.
Mark points if helpful.

We21 XML File Port definition

Hi All,
I want to define a port of type XML File in we21 of my R/3 system. Can I make it point to a non-local directory? (i.e. a directory that resides on a system on the network).
I tried to assign a Logical Directory in we21.I assigned an existing Logical File Path (T-code <b>FILE</b>) to a valid network path. But I get the error:
<i>Access to a stored directory not possible in this environment.</i>
How do I go about assigning a network path in we21?
Regards,
Puloma.

Hi,
Try this way
Define a Partner of type FILE
In WE21 under that you create a port for the partner FILE
Select physical directory and give the directory as
<XI Install dir>:\usr\sap\X3A\DVEBMGS31\data
and the file name in outbound file.
or any path from XI installation directory.
Once this is done try sending out the IDOC and see.
Thanks,
Prakash

EDI WE21 requrement Urgent...

Hi,
I have a requirement in the EDI,
I need to create a .txt file in AL11 transaction. For this I have configured in WE21 transaction and created a file port. In that I have chosen IDoc record type SAP Release 4.x. and Physical directory.
Given some path /usr/sap/interfaces/ in the Directory box.
In the Outbound file I have given RECON.TXT.
Now I have created the IDOC and the file is generated Recon.txt. I have a problem here. When ever the next IDOC is coming, it is overwriting. I have seen the function module which has been provided by SAP in WE21. But the client requirement is whenever first idoc is generated RECON1.txt should be created and when the next idoc is coming RECON2.txt shouldbe created in AL11. Other wise the file name should be Ordernumber_recon.txt.
If any one has gone trough the same scenario please through some lights on that.
Thanks and Regards,
Manohar

In WE21, you might have noticed there is a field called "Function Module" while defining "File" port. In this field we can define a function module to create a file name. Several function modules are available such as following
EDI_PATH_CREATE_CLIENT_DOCNUM     Directory + file name in format T_Client_Docnum
EDI_PATH_CREATE_DATE_TIME     Directory + file name in format T_CCYYMMDD_HHMMSS
EDI_PATH_CREATE_LENGTH_LE_8     Directory + file name in format NTHHMMSS.T
EDI_PATH_CREATE_USERNAME     Directory + file name in format SY-UNAME
However none of these will fulfill the requirement raised by client for filenames as RECON1.txt, RECON2.txt etc...
So i suggest you should copy a standard function module for example EDI_PATH_CREATE_DATE_TIME and modify its abap code to fulfill your needs. In your case you simply need a running counter (1,2,3...)as the first part of filename is constant (RECON). So in your copied function module you can use a customized Z table to store the latest value of counter or else you can use SAP number range object (you have to create a number range for that transaction SNRO and know how to call it in your abap code, normally with function module NUMBER_GET_NEXT)
After creating the function module remember to register it first in transaction WE55 and you will be able to use it in WE21.
I hope this will help you.

How to set up a virtual directory on Application Server

The environment of my application is Application server 9i version 2, on windows 2000, form 9.0.2.
I am trying to get the ORACLE 9i forms show_document to work.
I have successfully created an out file in the server directory.
I need the document to be displayed by the form.
Let me tell you more about my server.
I installed Oracle application server on drive E.
I copied my application to drive E and named PY which has 4 folders including :
E:\PY\fmb (for fmb)
E:\PY\forms (for fmx)
E:\PY\sql (for sql programs)
E:\PY\locrep (for lst files which are generated from sql programs)
Someone told me I should create a virtual directory, e.g. docs, in your httpd.conf file to point to e:\py\locrep
I've seen 4 httpd.conf files so I don't know which one I should modify and what should I type in exactly on what part ????
E:\ora9ias\apache\apache\conf\httpd.conf
E:\ora9ias\apache\apache\conf\httpd.conf.default
E:\ora9ias\apache\apache\conf\httpd.conf.smibak
E:\ora9ias\dcm\config\plugins\apache\httpd.conf
After I modify the httpd.conf file, so what will my URL be ? Also, I have a lot of reports, I would like to run different reports at the same time ??????
Thank you very much.
Trang

The best way to edit httpd.conf is using Enterprise Manager Website. Open this site (http://machine:1810) and drill down on your application server, and then on HTTP Server. Open link Advanced server Properties. Now you see a list of all configuration files. Click on a file to open it.
To add the virtual path:
- You could best add your virtual path in file forms90.conf (it is included in httpd.conf with the <include> tag).
- Within this file you already find some virtual path definitions: Look for the AliasMatch tag, e.g. for the 'forms90/html' virtual directory.
- Copy this line and change 'html' into the directory name you like to use, e.g. 'myfiles'. Map it to the physical directory (on your E: drive) where your files are. (E: must not be a network drive, however).
- browse to your files using: http://machine:7777/forms90/myfiles/filename.
Peter

How to create a virtual directory for the Report Server

Hi, I am new to server admin and need help with configuration.
I would like to map a physical directory on the server so that users can use a virtual path to save report outputs in that location. For example the form will call the report with desname = 'd:\xyz\abc.pdf' and the output file will be saved in a predefined directory on the server. I know how to do it on the forms server using the orion_web.xml but not sure about the report server configuration.
We are using rwservlet to run reports on 10G release2/Windows 2000.
Thanks in advance,

Virtual directories are mapped in the httpd.conf file in the Apache directory as aliases:
Alias /xyz/ "d:\xyz\"
You can now call the report output with
http://server/xyz/abc.pdf

LSMW - Logical File Path not pointing to Physical Path

Hi All,
I'm trying to upload some MIGO goods movement transactions through LSMW. I'm to step 7, specifying files, and am getting the error "No logical file path has been specified. I saw another thread which helped me, but I am still stuck after reading it.
Re: Error in Creation of Purchase Requisition through LSMW
I have gone into transaction FILE and created the logical file name MM_GOODS_MOVEMENT, and set the physical file as the .lsmw.conv converted data file as the above thread says. I also tried using the logical path LOCAL_TEMPORARY_FILES for the converted data, but I get the error telling me that the path does not point to a physical directory. I then tried creating my own file path, and setting the path to my desktop where the source file is, and the same error occurred.
Does anyone know why it is not recognizing the physical path maintained in the logical path?
Thanks,
J

Thanks for both replies,
I have full authorization across all transactions in the sandbox system I am working in, so that is not an issue. I filled the logical path field in the MM_GOODS_MOVEMENT logical file with LOCAL_TEMPORARY_FILES, as well as trying my own created file path, and I recieved the same error for each. Here is how the config is currently:
Logical File Path Definition: MM_GOODS_MOVEMENT
Syntax Group: WINDOWS NT
Physical path: C:\Documents and Settings\jchanski\Desktop\<FILENAME>
or,
Logical File Path Definition: LOCAL_TEMPORARY_FILES
Syntax Group: WINDOWS NT
Physical path: C:\temp\<FILENAME>
Logical File Name Definition: MM_GOODS_MOVEMENT
Physical file: MM_MM_MIGO_INVUPLOAD.lsmw.conv
Data format: ASC
Applicat.area: MM
Logical path: LOCAL_TEMPORARY_FILES or MM_GOODS_MOVEMENT
Do you see any error with this Brajvir? Thanks!

FTP Adapter on OSB - how to specify logical directory path

I am using the ftp adapter on osb 11g (Oracle Service Bus Version: [Oracle Service Bus Server Side Dependencies 11.1 Thu Aug 19 02:10:08 PDT 2010 ] Oracle Weblogic Server Version: [WebLogic Server 10.3.4.0 Fri Dec 17 20:47:33 PST 2010 1384255 ])
I defined the ftp adapter using jdev and imported the jca artifacts into osb.
I created a business service using jca adapter to put the file to the ftp server.
The problem is the write directory on the ftp server is different for each environment - DEV, QA, PROD.
I used a physical directory specification when defining the adapter in JDEV because it wasn't clear from the help how i would override a logical directory during deployment to osb (it is clear if deploying to soa suite).
Is there a way to override this dynamically override this directory at runtime. Or even better a way to override during deployment time?

You do it either way....
Import the SBConfig jar into the sbconsole and change the value in the JCA file
or
Re: Customizing JCA FTP Adapter properties in OSB

Changing File Adapter target directory - container bounce required?

I edited USPSshipment.wsdl in the FulfillmentESB project to reflect my desired target directory (/home/oracle/usps_ship instead of C:\temp) and then re-registered the project with ESB.
The registration completed normally, and I verified that the edited USPSshipment.wsdl file exists in the deploy and content directories in the ESB home. Also, checking the WSDL file from the ESB console showed the new value.
However, until I bounced the OC4J_soa container, order submissions continued to use the C:\Temp target. After I bounced the container, the next order submission went to the desired /home/oracle/usps_ship.
What is the expected cacheing behavior here? Is there a programmatic way to clear the cache after an ESB deploy. It is too restrictive if I have to bounce the whole container serving ESB and BPEL (perhaps many processes and ESB adapters) just to make a change in one particular project.
Thanks.

Using the SOA Demo, FulfillementESB project ~
Example: Immediate reflection of change w/o further action
If I change the Shipment router filter criteria (Fulfillment_Shipment.esbsvc) to cut at say $750 instead of $500, then re-register FulfillmentESB with OrderBookingIS, JDeveloper reports:
Registration of Services Successful
Fulfillment Shipment Updates
The next submitted order follows the new criteria (e.g. a $600 order goes out USPS).
Example: Change not reflected in runtime, until OC4J container restarted
If I change the target physical directory of the USPS Shipment file adapter (USPSShipment.wsdl), then re-register FulfillmentESB with OrderBookingIS, JDeveloper reports:
Registration of Services Successful
However, the next submitted orders continue to be written to the previous target physical directory. Checking the value of the re-deployed USPSShipment.wsdl at the ESB host filesystem and through the ESB console shows the updated value, but the old copy is being actually used to process the message.
Restarting the OC4J container, causes the updated USPSShipment.wsdl to be referenced for messages submitted thereafter.
-Todd
Message was edited by:
tbeets

Physical Directory in WE21

Similar Messages

Maybe you are looking for