Java + PL/SQL modules w one login

We have a huge (600+ modules) collection of applications done in web pl/sql via Designer. We are moving to JDeveloper for future work and would like our end-users to be able to log into each individual application just once regardless of whether its pl/sql, java or a mixture. Implementing SSO isn't feasible at this time because the pl/sql apps use user for the audit columns, and there are literally hundreds of tables that would need their apis fixed.
I can get my java apps in stand-alone bc4j to use the same realm as the pl/sql, but when I import those applications to into 10AS the application server seems to set the realm and the url prefix to the same value, and since mod_plsql sees the request first nothing gets to the java apps.
My other issue is that getRemoteUser shows the realm name in front of the username e.g. testapp/testuser and the HTTP_REMOTE_USER variable on the pl/sql side shows only the username.
Perhaps I'm approaching this the wrong way so any advice on how to consolidate logins while keeping the app-user is database-user configuration on the pl/sql side would be welcome.

One possibility is to use a Java servlet that replaces mod_plsql. Here is a link to a commercial one called JOPA: http://www.dynamicpsp.com/dpsp/prod/!go?ln=p_jopa_wp. Or there is the open source DBPrism servlet (http://www.dbprism.com.ar/dbprism/doc/Home.html) - which is primarily for getting XML out of the database, but also can act as a replacement for mod_plsql. Warning- I have not used either product and cannot promise that either is actually useful.

Similar Messages

Does Java support mutliple sql statements in one call?

statment.executeUpdate("DROP DATABASE IF EXISTS diy55;CREATE DATABASE IF NOT EXISTS diy55 DEFAULT CHARACTER SET utf8;GRANT SELECT,INSERT,UPDATE,DELETE,CREATE,DROP,ALTER,INDEX,USAGE on diy55.* TO 'diy55'@'localhost' identified by 'diy55';FLUSH PRIVILEGES;");Java seemes doesn't support multiple sql statements in one call, just like PHP mysql_query(), is it?

Then just extend the java.sql.Statement to your own Statement class and write logic which splits the given query at the right places and executes them one by one.
That the mysql_query() already has it built-in wouldn't say that it isn't possible in Java. Comparing with PHP is a bit pointless, it is a completely different language.

Problem in JAVA PL/SQL Block

Hi all,
I am facing a peculiar problem. I have one JAVA PL/SQL code, which runs a shell script, which indeed fire a zipping command remotely. Here I am providing you the code. Actually it zip some files at application server running from Database Server.
create or replace procedure RC(p_cmd in varchar2)
as
x number;
begin
x := run_cmd(p_cmd);
EXCEPTION
when others then
dbms_output.put_line('WHEN OTHERS '||sqlcode||sqlerrm);
end;
<====This RC is called from a before insert trigger. And value of p_cmd is $ORACLE_BASE/report_compress.sh $REPORT_DIR/<report_file>.
create or replace
function RUN_CMD(p_cmd in varchar2) return number
as
language java
name 'Util.RunThis(java.lang.String) return integer';
create or replace and compile
java source named "Util"
as
import java.io.*;
import java.lang.*;
public class Util extends Object
public static int RunThis(String args)
Runtime rt = Runtime.getRuntime();
int rc = -1;
try
Process p = rt.exec(args);
int bufSize = 4096;
BufferedInputStream bis =
new BufferedInputStream(p.getInputStream(), bufSize);
int len;
byte buffer[] = new byte[bufSize];
// Echo back what the program spit out
while ((len = bis.read(buffer, 0, bufSize)) != -1)
System.out.write(buffer, 0, len);
rc = p.waitFor();
catch (Exception e)
e.printStackTrace();
rc = -1;
finally
return rc;
And also I have executed:
BEGIN
dbms_java.grant_permission('<DB_USER>','java.io.FilePermission','$ORACLE_BASE','write');
dbms_java.grant_permission('<DB_USER>','java.io.FilePermission','$ORACLE_BASE/report_compress.sh','execute');
dbms_java.grant_permission('<DB_USER>','java.lang.RuntimePermission','*', 'writeFileDescriptor' );
END;
REPORT_COMPRESS.SH
#!/bin/sh
remsh <APPLICATION SERVER IP> -l <APPLICATION_SERVER_OS_USER> -n "gzip -f $1"
Now problem is while execute RC, it gives no error but the remote file does not get zipped. When I run report_compress.sh from command promt, it works fine.
Earlier we used HP-UX as our DB server and Application Server. And RC worked fine there. Recently we have moved to AIX(for both DB and AS), and now we are facing this problem.
Please guide me to resolve this problem.
Note: .rhosts entry is fine in both the servers(as per me) as report_compress.sh is working fine.

Make sure that the .profile does not read something from tty - it should not, but it sounds like the .profile is expecting tty input.
Alternatively, simply copy and paste the environment settings from .profile to your compress script. The only read important setting is PATH as your script simply does a remote shell. Thus LD_LIBRARY_PATH/SHLIB_PATH and other settings are not needed.
You can also add debugging to your shell script - useful when executing it in the background.
Something as follows:
# create debug file name
DEBUGFILE=/tmp/~$$.debug-out # e.g. /tmp/~1234.debug-out where 1234 is the PID
echo "`date` Starting [$0] with parameters [$*]" >> $DEBUGFILE
# sourcing environment
echo "Loading default profile..." >>$DEBUGFILE
. /home/oracle/.profile 1>>$DEBUGFILE 2>>$DEBUGFILE
# running remote shell
echo "Running remote shell..." >>$DEBUGFILE
remsh ... 1>>$DEBUGFILE 2>>$DEBUGFILE
.. etc.
When this is working, you can remove the debug file from the shell script, or simply delete it when the script is successful.

One login for multiple applications

Hi,
I was using HTMLDB 1.6 and followed the following guidlines: one login multiple applications for one login to multiple applications. Basically I set the cookie name the same for both applications. It worked.
Now I've upgraded to HTMLDB 2.0 and this doesn't work anymore. I need to login to each application. Any way around this?
Thanks,
Marty

Scott
I have an SQL report region that includes this column:
decode((select count(*) from gr_agency_add f, gr_emp_add a where a.empid = g.userid and a.aid = f.aid and a.add_type = g.address_option), 0, '', '<a href="f?p=&APP_ID.:9:' || :APP_SESSION || '::::P9_GRID:' || g.GRID || ' target="_blank">') ||decode(trunc(length(g.topic)/30), 0 , g.topic || '</a>', substr(g.TOPIC,1,30)||'..') "Topic",
With the
target="_blank"
of course I am opening page 9 in a new window. I never had a problem with this but today, in a demo to some executives, the new window apparently came up with the login page and not page 9. Would this be due to a browser setting? Unfortunately, I don't have much info about the laptop they were using.
The demo was still a success!
Bill

Can't connect to SQL Database using new login

I can connect to Azure SQL Database using the admin ID created when I set up the Azure account. Now I want to set up a different login that has only regular read/write privileges. I have followed the instructions on several web posts, to create
a login at the server level, then create a user at the database level from the login. When I try to connect, though, the connection fails. I get an error message, and a GUID that I am supposed to supply to MS tech support, except that all I have
is a standard Azure subscription, which does not include tech support.
Here is what I have done so far:
Logging in to the server with the admin account with SQL Mgmt Studio, I go to the master database on my Azure server, and create the login:
CREATE LOGIN clouduser@gxw8x04nlb with PASSWORD = 'xoxoxoxo';
This appears to work (password supplied is a "strong" password), or at least it doesn't show any error response.
Then, connecting to the production database in Azure, I set up a user and set permissions as follow:
CREATE USER clouduser1 FROM LOGIN clouduser@gxw8x04nlb;
EXEC sp_addrolemember 'db_datareader', 'clouduser1'
EXEC sp_addrolemember 'db_datawriter', 'clouduser1'
When I attempt to connect to Azure using the new login, however, I get an error message (from SQL Mgmt Studio as well as from MS Access). I am using the SQL Server Native Client 10.0.
The message is:
Connection failed:
SQLState '28000'
SQL Server Error 18456
[Microsoft][SQL Server Native Client 10.0][SQL Server]Login failed for user 'clouduser'.
Connection failed:
SQLState: '01000'
SQL Server Error: 40608
[Microsoft][SQL Server Native Client 10.0][SQL Server]This session has been assigned a tracing ID of
'271851d5-8e94-497c-a332-d9d40682bb7a'. Provide this tracing ID to customer support when you need assistance.
Is there some missing step I need to do, to permit the login to see the database? I have looked for such in the various discussions about connection problems, but have not been able to find such a thing. Or extend more permissions to the new
user? Or specify a default user ID for a given login ? (as they do not have the same names, in the examples I have seen). I tried making the user ID the same as the login (w/o the server name), but that didn't seem to help, either.
I have done numerous web searches, and tried about every variation of login or user ID, password, etc. that I can think of, and all of them encounter the same error. It's got to be something very simple - clearly Azure supports more than one login
per database. But's that's all I have at the moment. That login connects just fine, but others won't, using the same PC, middleware, IP address, etc.
Any help would be much appreciated -
Thanks,
Doug

Olaf -
I noticed that, but all of the examples I have seen have Users named slightly differently from the Logins. I did try logging in with the Login name instead. No good. I tried making the Login and the User name the same (clouduser). Also
no good - same symptoms.
I think the problem is that the initial creation of the Login needs to be done WITHOUT the server name on the end.
Logged in to the master DB, I tried: Create LOGIN clouduser@gxw8x04nlb with PASSWORD = 'xoxoxoxo';
That didn't give me an error message, but I think it created a LOGIN of 'clouduser@gxw8x04nlb'. When
referenced from the outside world, I would need to specify 'clouduser@gxw8x04nlb@gxw8x04nlb'.
I tried deleting the logins, and then creating the login 'clouduser' in the Master DB. Then in
the application DB, I created User clouduser from LOGIN clouduser, and then assigned a role. That seems to work!
In my connection string, File DSN, etc. I still need to supply the login as 'clouduser@gxw8x04nlb'.
But when I am logged in to the server, and working in the Master DB or the application DB, I just refer to the Login as 'clouduser'.
Seems a little more complicated than it really should be, but at least I now have something that works.
Doug
Doug Hudson

Add multiple users to a certain database in the SQL server in one go

I wonder if there is any method to add multiple users to a certain database in the SQL server in one go and without using transact code.
I can add a single user to a certain database “Q” for example by right click on the user “U1” for example then properties then in user mapping tab select
the database “Q”, so there should be a method to add multiple users ( U1,U2,U3…) to this database?
Best Regards,
Ahmad

Many thanks Visakh16,
I can do this using the below script, but what I am searching for is do to this without any scripting.
USE TestDatabase; --Make sure you have the right database
DECLARE @sql VARCHAR(MAX) = '';
SELECT @sql = @sql + 'CREATE USER ' + name + ' FOR LOGIN ' + name + ';
' +
'GRANT CREATE TABLE, CREATE PROCEDURE, CREATE VIEW, VIEW DEFINITION TO ' + name + ';
FROM sys.server_principals p
WHERE p.type in ('S','U') -- SQL Logins and Windows Login. Do not change!
and p.name in ('U1','U2','U3'); -- List of names to add. alter to suit
PRINT @sql; -- Show the statements being executed in the messages pane
EXEC(@sql); -- Run the statements that have been built
Thanks,
Ahmad

Reporting on PL/SQL module dependencies

Hi,
I've just reverse-engineered a database into Designer 10g. The database contains 30+ packages and assorted other stored PL/SQL code modules (functions, procedures & triggers). I have used the Dependency Manager to build a list of dependencies for each code module, so now I can browse through each procedure/function and see which modules are called.
I would like to be able to generate a paper report of all the PL/SQL module depoendencies. Something on the lines of :
Package Name
....Procedure1
........Calls proc2
........Calls func3
........Calls package.proc4
....Function1
........Calls ...
Does anyone know a way to generate such a report of PL/SQL module dependencies from within Designer?
Thanks in advance
Steve.

You can get some of what you want with the Module Network report (cdmodnet) found under Module Design in the Repository Reports. Make the Module Sub-type "PL/SQL Definition". But if this doesn't do what you want, you may need to write your own report. This report would be a good starting place, and the Repository Reports tool gives you a way to copy an existing report to write a new one. Look at the queries, and that will tell you which API views to query and how to join them.

Inconsistent java and sql object types

Hi,
I have run into error "Inconsistent java and sql object types"
while mapping a java class to a sql object type. The java class
is just a duplicate of sql data structure and I pretty much
follow the JDBC Developer's GUide's examples (20-43 to 20-45)
to create the mapping java class.
Any one runs into simliar problem or any clues?
Thanks,
Ed
Exception in thread "main" java.sql.SQLException: Inconsistent java and sql object types: InstantiationException:
at oracle.jdbc.dbaccess.DBError.throwSqlException(DBError.java:168)
at oracle.jdbc.dbaccess.DBError.throwSqlException(DBError.java:210)
at oracle.sql.STRUCT.toClass(STRUCT.java:433)
at oracle.sql.STRUCT.toJdbc(STRUCT.java:366)
at oracle.jdbc.oracore.OracleTypeUPT.unpickle80rec
(OracleTypeUPT.java:236)
at
oracle.jdbc.oracore.OracleTypeCOLLECTION.unpickle80rec_elems
(OracleTypeCOLLECTION.java:553)
at oracle.jdbc.oracore.OracleTypeCOLLECTION.unpickle80rec
(OracleTypeCOLLECTION.java:383)
at oracle.jdbc.oracore.OracleTypeCOLLECTION.unpickle80
(OracleTypeCOLLECTION.java:329)
at oracle.jdbc.oracore.OracleTypeCOLLECTION.unlinearize
(OracleTypeCOLLECTION.java:218)
at oracle.sql.ArrayDescriptor.toJavaArray
(ArrayDescriptor.java:501)
at oracle.sql.ARRAY.getArray(ARRAY.java:197)

The safest way would be to use JPublisher to generate the type classes. In your application, you can just use the generated code to manipulate the object.

ERROR: Inconsistent java and sql object types

I am getting a nested object table in the form of an oracle.sql.ARRAY. I then call getResultSet() on this array and try to cast each object into a Java implementation of the SQLData interface (also defined in the connections type map).
This approach works for one nested object table but not for another. I get the error message:
"Inconsistent java and sql object types"
Does this mean that my Java to Oracle type mappings are wrong? Any help would be appreciated.
Regards
Steve

i had a similar situation. found out it was occuring because the oracle type had not been granted to the appropriate user.
todd kegley

Java + Open SQL sample code..

Hi All,
I want to use Open SQL in our project, Can any one send me a Stand alone sample example of Java and Open SQL.
also point me, from where i can get informations regardig the same..
We have SAP J2EE 6.20.
Also pls let me know the supporting jar,zip files, and from where can i download the same.
I googled, but i couldnt get any sample examples..
1. Does SAP J2EE 6.20 support Java & Open SQL.
2. What all supporting jar files, do i need and from where can i get the same..
Regards,
Shridhar..

Hi Sridhar,
           To use open sql in particular project in Netweaver Studio,you should have WAS 6.40.SO the current version which you had mentioned WAS 6.20 doesn't support open sql.
           Secondly to use open sql you should place the driver .jar file in the lib folder of your respective perspective.you can download the same .jar file from the vendor's website(eg:- oracle).Then you need to conect to the DB,have youe executable statement and iterate thro' your resultset similarly what you in an ordinary JDBC program.
            Hope this helps.
Regards,
Guru

How to keep multiple function modules under one Web service

Hi Experts,
I have Three RFC function modules and i need to create one web service for these three RFC function modules. I know How to crearte a web service for one function module.
please suggest me How to keep multiple function modules under one Web service.
Thanks in advance
Lakshminarayana

Hi Lakshmi,
The best way to do it is to assign all the three RFC Enabled FM's to one function group. Later on the top menu in Utilities you get an option to Create a Webservice from a Function Group.
You can create one single Webservice using all the the 3 FM's.
I hope this helps.
Thanks,
Manu

Crash Affecting only One Login - But Not in Safe Mode

I have an iMac (built-in iSight) that suddenly has problems with only ONE login (the other accounts on the machine work fine). The troubled login won't mount the desktop at all, even with the entire library folder removed. Ran DW utility, but that didn't help. The interesting thing is, if I boot in SAFE mode, the problem goes away.
What am I missing? If there's a problem with some bit of startup code, driver, etc., why is it only affecting one login?
I realize my info above is sketchy. Hoping it might be enough for somebody to offer a clue... thanks.

Hi, Lonestar.
An errant Startup or Login Item maybe the cause. In this case, it would likely be a Login Item defined in the affected account. My "Troubleshooting Startup and Login Items" FAQ can help you pin that down if such an item is causing the problem. Focus on the instructions concerning Login Items with the affected account.
Good luck!
Dr. Smoke
Author: Troubleshooting Mac® OS X
Note: The information provided in the link(s) above is freely available. However, because I own The X Lab™, a commercial Web site to which some of these links point, the Apple Discussions Terms of Use require I include the following disclosure statement with this post:
I may receive some form of compensation, financial or otherwise, from my recommendation or link.

How do I find out if Sql server Agent's login has rights to a DB?

How do I find out if the SQL server Agent's login has rights to the target DB, that is the DB on which I will be executing a query ? Is there a query for this ?
Thanks.

The job will run with the permissions of owner of the job and only be modified by the owner
or sysadmin role members.
you need get the owner of the job and for example 'ABC' is the owner of the job then pass value in the below query
DECLARE @DBuser_sql VARCHAR(4000)
DECLARE @DBuser_table TABLE (DBName VARCHAR(200), UserName VARCHAR(250), LoginType VARCHAR(500), AssociatedRole VARCHAR(200))
SET @DBuser_sql='
SELECT ''?'' AS DBName,a.name AS Name,
a.type_desc AS LoginType,
USER_NAME(b.role_principal_id) AS AssociatedRole
FROM ?.sys.database_principals a
LEFT OUTER JOIN ?.sys.database_role_members b ON a.principal_id=b.member_principal_id
WHERE a.sid NOT IN (0x01,0x00) AND a.sid IS NOT NULL AND a.type NOT IN (''C'')
AND a.is_fixed_role <> 1 AND a.name NOT LIKE ''##%'' AND
''?'' NOT IN (''master'',''msdb'',''model'',''tempdb'')
ORDER BY Name'
INSERT @DBuser_table
EXEC sp_MSforeachdb @command1=@dbuser_sql
SELECT * FROM @DBuser_table ORDER BY DBName
--SELECT * FROM @DBuser_table where username='ABC' ORDER BY DBName
Useful links
http://www.mssqltips.com/sqlservertip/2317/running-sql-server-agent-with-a-least-privilege-service-account/
http://www.brentozar.com/blitz/jobs-owned-by-user-accounts/
-Prashanth

Copying function module from one function group to another

What is the efficient way to copy a module from one group to other so that all its subroutines and any dependencies with in the FM can also be copied?

Hello,
Tcode:SE37
Menu: FunctionModule->OtherFunctions->Copy in that there are different option give a new name of the function group this should work
regards
suresh nair

No Un-ambiguous Selection of Mapping Module (above one) for BUPA_MAIN BDocs

Hi All
I seem to have hit a snag here. I am trying to replicate customers from CRM to ERP but am getting these two errors:
Main error: Partially send, receivers have errors.
No un-ambiguous selection of mapping module (above one) for BDoc type 3932399B34D6279BE10000000A11449E.
Service that caused the error: SMW3OUTBOUNDADP_CALLADAPTERS._
I have searched on Googleand SDN but have come up with nothing as far as these errors are concerned. We have managed to replicate Business Partners, Materials and Conditions from ERP to CRM without any problems but when it cam e to replicating back, these errors started popping up.
I have used all the tools for analysis but have come up with nothing the really suggest to me what is wrong. I have monitored queues, traces and dumps. I have played around with the filter setting and all.
With regards to the process itself, i followed the C03 best practices config guide and i believe that i have everything in place. Here are the steps i followed:
1.Create an Internal number range for the CRM customer group.
2. Assigned the number range to the Customer group
3. Created a corresponding number range (External) in ERP and a corresponding CRM Customer account group as a copy of the standard 0001 Sold-to-party account group and assigned the number range to it.
4. Defined Partner functions (SP, SH, BP, and PY) for the account group
5. I mapped the CRM classification for customers (B) to the created account group (CRM Customer)
6. I synchrponized the fields for the account group created and the BP roles in CRM including changes on a per client (BUPA) basis.
7. I redowloaded all the cutomizing objects (though i only needed the DNL_CUST_KTOKD) for the sake of the mappings directly in CRM or R/3 Account group maintenance during the creation of the BP.
8. I customized the filter settings in BUPA_MAIN and included the BUT100 table for the filetring of the customers who were created as SP, BP, PY, or SHand who belonged to the Customer group we had created. (BUT000->BU_GROUP & BUT100->RLTYP) for source site CRM.
9. Activated the necessary function modules through transactions COM_BUPA_CALL_FU in ERP and CRMC_BUT_CALL_FU in CRM according to [http://help.sap.com/saphelp_crmscen70/helpdata/en/60/85b5d333f4174b8e982c1c15db35f3/frameset.htm]
10. Using transaction R3AS i sought to replicate the BUPA_MAIN object.
Because of the few customers created in CRM, it only executes within a single block so the queues would be empty by the time SMQ1 and SMQ2 are executed. R3AM1 shows it as running. SMW01 shows the above error.
When creating a BP through the tx code BP, the main error would be Technical error(Incomplete) and the two other errors as mentionme above would also be there
Is there anyone out there who has met this very same problem or knows the solution to this.
Thanks in advance.

We managed to get a solution to the problem. Here it goes:
When a new adapter object (e.g a custom one) is created, the SMOFOBJECT table is populated with the entry and, in turn, the SMOFUPLMAP is populated with the object and the mapping module for uploading to R/3. During the upload process, a search for custom mapping modules is done and if not found the standard mapping modules are used. However, and it seems weird since it seems as if it applies to the BUPA_MAIN object or the corresponding mapping module (BUPA_MWX_BDOC_UPLOAD_MAIN_R3), there has to be a one-to-one mapping between an mapping module and adapter object.
I had created a custom adapter object (Z_BUPA_MAIN) to which I wanted to put some custom filters and did not want to test with the standard object. So two objects using the same mapping module ended up existing in the SMOFUPLMAP table (one-to-many mapping). The function module in which this happens is the GET_UPL_FUNCTION_FOR_OBJ.
So the solution would be to delete the object if it is not needed or to maintain the SMOFUPLMAP table to remove the entry. When you delete the object, it sometimes does not automatically/subsequently remove the entry in SMOFUPLMAP, so manual maintenance has to be done or the same object has to be recreated and have the mapping module allocated. After this, the problem would be solved. What is weird to me is that the only one-to-one mapping is allowed. I created a custom object for the download of conditions and I did not get that error. So it means that in this case you would have to copy the mapping module and create a custom one for the custom object.

Java + PL/SQL modules w one login

Similar Messages

Maybe you are looking for