Monday, 25 June 2012

How to create and export a report of all Service Requests ( SR’s)



Navigate to the "My Oracle Support" web site "https://support.oracle.com" and log-in.

From the top of the web page, click on the "Service Request" tab.

If you wish to include closed SR's in your report, click off the "Include Hard Closed" check box on the top right-hand corner of the "Service Request" window.
  
Click on the "Excel Spreadsheet" icon to the left of the "Service Request" window and select "Columns" to select of deselect fields. Select the "Printable view" option to view and print the report.

Note: We can also use : Click on Excel spreadsheet icon > copy to clipboard > all rows > paste in a desired location ( example: MS excel).








































Reference Metalink note: 1058974.1

Tuesday, 19 June 2012

ORA-00257: archiver error. Connect internal only, until freed


While trying to connect to the database, we get the below mentioned error:

ERROR:
ORA-00257: archiver error. Connect internal only, until freed.

Cause:


-       The error is caused by the fact that the database archiver process is unable to archive the current online redolog due to lack of space in the destination for the archivelogs.

-       The flash_recovery_area is full.

Applicable to 9i, 10g and 11g database versions.

Solution:

The basic idea is to free space so as to make room for new archiving to happen. This can be addressed in multiple ways.

-       Allow more space in the DB_RECOVERY_FILE_DEST with the DB_RECOVERY_FILE_DEST_SIZE parameter :

                                           alter system set db_recovery_file_dest_size=5G ;

-       Imagine a situation where the space increased in step 1 gets filled, ie, 5 GB. Set an alternative archiving destination, so that when the primary destination is full, archiving is automatically  performed to the alternate destination:

                            Following changes in parameter file can enable the same:

log_archive_dest_1='LOCATION=use_db_recovery_file_dest NOREOPEN ALTERNATE=LOG_ARCHIVE_DEST_2'
log_archive_dest_2='LOCATION=/other_dest_for_archiving'
log_archive_dest_state_1='enable'
log_archive_dest_state_2='alternate'
db_recovery_file_dest='/db01/mine/oracle/product/10.1.0/db_1/flash_recovery_area'
db_recovery_file_dest_size=5G



-       Check the archive_log_dest  and verify if the path mentioned exists and is valid.

-       Take a backup of the existing archive files (recommended) OR it can be deleted to make more free archiving space.

This can be done at the server level where files can be manually deleted. However incase of ASM storage with rman, below mentioned way is recommended (which would perform backup and then deleting):

                                  Example :

                                rman target sys/sys
                                 RMAN> backup archive log all ;
                                 RMAN> delete archive until time 'trunc(sysdate)';

Please refer : http://docs.oracle.com/cd/B19306_01/backup.102/b14192/bkup003.htm  for more details on RMAN archiving.

Thursday, 14 June 2012

Concurrent managers are down with status: “System Hold, Fix Manager”



Issue:

All the concurrent managers are down with a status message: system hold, fix manager.

The administer Concurrent manager screen has Actual: 0 and Target: 1 for all managers.



Solution:           Ensure all application services are down.

     -  Please check the system profile: Concurrent: GSM Enabled. This should be set to ‘Y’

     -  Execute: afdcm037.sql Path : $FND_TOP/patch/115/sql

Description of the script:

 PL/SQL script to create libraries for FNDSM and FNDSVC and create Managers for Preexisting Nodes.  Future nodes will be handled by db trigger.

 Relink the below mentioned executables:
                                              
                                           adrelink.sh force=y “fnd FNDLIBR”
                                           adrelink.sh force=y “fnd FNDSM”
                                           adrelink.sh force=y “fnd FNDFS”
                                           adrelink.sh force=y “fnd FNDCRM”

   Execute cmclean.sql to clean out the concurrent manager tables.


 Start all the application services and restest the issue.

Tuesday, 12 June 2012

Move Concurrent Processing Server from one node to another in Oracle Apps 11i

Example:

Oracle E-Business Suite (11.5.10.2) with 2 nodes:

Node 1: Database and the Concurrent Processing
Node 2: Forms, Reports and Web Server.


Goal: To Move the concurrent processing from Node1 to node 2.


Requirement: Shared appl_top / Shared application tier file system. If not follow the Step 2 below.

How to perform:

1. This is recommended to be performed on a NON-PRODUCTION  system first with necessary backups.

2. Create a shared APPL_TOP by following Metalink Note 233428.1 for sharing the application tier files
system and allow the node 2 to share for concurrent manager node ( Node 1).

3. Edit context file on the Forms and Web node2 and change following context variables:

s_isConc from "NO" to " YES "
s_concstatus from "disabled" to " enabled "

On Node1: make the below changes

s_isConc from "YES" to "NO"
s_concstatus from "enabled" to "disabled"

4. Run autoconfig on all the APPS nodes.



Reference:

Metalink notes:  373611.1 and 260887.1.