TriniDBA

Problem: After cloning the production EBS instance, user submits the Subledger Period Close Exceptions  Report, but it does not complete.

Solution: Post-processing of the request is points to the production instance (for some reason). Run AutoConfig to point it back to the clone.

Checking the log file of the concurrent request shows:

+---------------------------------------------------------------------------+
Subledger Accounting: Version : 12.0.0

Copyright (c) 1979, 1999, Oracle Corporation. All rights reserved.

XLAPEXRPT module: Subledger Period Close Exceptions Report
+---------------------------------------------------------------------------+

Current system time is 06-OCT-2014 08:55:34

+---------------------------------------------------------------------------+

XDO Data Engine Version No: 5.6.3
Resp: 50263
Org ID : 82
Request ID: 2001850
All Parameters: P_APPLICATION_ID=707:P_JE_SOURCE=Cost Management:P_LEDGER_ID=2025:P_LEDGER=Comp-Ledger:P_PERIOD_FROM=AUG-14:P_PERIOD_TO=AUG-14:P_DUMMY_PARAM_1=1:P_EVENT_CLASS_CODE=:P_EVENT_CLASS=:P_DUMMY_PARAM_2=1:P_JE_CATEGORY_NAME=:P_JE_CATEGORY=:P_MODE=S
Data Template Code: XLAPEXRPT
Data Template Application Short Name: XLA
Debug Flag: N
{P_JE_CATEGORY=, P_EVENT_CLASS=, P_JE_CATEGORY_NAME=, P_PERIOD_TO=AUG-14, P_MODE=S, P_DUMMY_PARAM_2=1, P_DUMMY_PARAM_1=1, P_EVENT_CLASS_CODE=, P_LEDGER_ID=2025, P_PERIOD_FROM=AUG-14, P_APPLICATION_ID=707, P_LEDGER=Comp-Ledger, P_JE_SOURCE=Cost Management}
Calling XDO Data Engine...
+---------------------------------------------------------------------------+
Start of log messages from FND_FILE
+---------------------------------------------------------------------------+
+---------------------------------------------------------------------------+
End of log messages from FND_FILE
+---------------------------------------------------------------------------+

+---------------------------------------------------------------------------+
Executing request completion options...

Output file size:
10435

+------------- 1) PUBLISH -------------+
Beginning post-processing of request 2001850 on node APPLE at 06-OCT-2014 08:55:38.

It appeared that the post processing was pointing to the apple server. To resolve this issue, I shut down the application tier services using the following commands:

cd /u01/test/inst/apps/TEST_six/admin/scripts
./adstpall.sh apps/apps

Sometimes after shutting down, there are lingering Oracle processes, especially concurrent manager processes. To ensure all processes are killed, I shut down the database tier processes as well using:

cd /u01/test/db/tech_st/11.2.0/appsutil/scripts/TEST_six/
./addbctl.sh stop immediate
./addlnctl.sh stop TEST

From there, you must ensure that all the Oracle processes have stopped. Check the processes using the command:

ps -ef | grep username

The output should appear as follows:However, after shutting down all the Oracle processes, there may still be lingering processes, such as:

• /u01/test/apps/tech_st/10.1.3/appsutil/jdk/bin
• /u01/test/apps/apps_st/appl/fnd/12.0.0
• FNDLIBR FND CPMGR FNDCPMBR sysmgr=”” sleep=30 pmon=4 quesiz=1 diag=N logfile=/d
• FNDLIBR FND Concurrent_Processor MANAGE OLOGIN=”APPS/ZG87987FB3D865184E213C16EB
• PALIBR FND Concurrent_Processor MANAGE OLOGIN=”APPS/ZGD810746A874C98D2F4CFDE8BA

Use the UNIX ‘kill’ command to kill these processes, so that the output of the ‘ps -ef | grep vision’ command appears as above.

From here, start the database tier services:

cd /u01/test/db/tech_st/11.2.0/appsutil/scripts/TEST_six/
./addlnctl.sh start TEST
./addbctl.sh start

Then run AutoConfig on the application tier:

cd /u01/test/inst/apps/TEST_six/admin/scripts
./adautocfg.sh

Enter the apps password when prompted, and the script will complete in 1-2 minutes. Then restart the application tier services:

cd /u01/test/inst/apps/TEST_six/admin/scripts
./adstrtal.sh apps/apps

Once the application tier processes have been restarted, you may advise the user to log on and retest the issue.