Fixing HTTP Service Issue After Sudden Power Shutdown of Oracle EBS 12.2 Application Server

 

Overview

Recently, I faced an issue in an Oracle E-Business Suite 12.2 environment after a sudden power shutdown of the application server machine.

After the server was powered on again and the EBS services were checked, I noticed that the HTTP service was not coming up properly. The rest of the environment appeared normal, but the web access was impacted because the Oracle HTTP Server service was showing as stopped.

This kind of issue can occur when the application server is not shut down gracefully. During an abrupt shutdown, some runtime or state files related to OPMN/OHS may remain in an inconsistent state, which can prevent the HTTP service from starting properly.

 

Environment

Application : Oracle E-Business Suite R12.2
Component   : Oracle HTTP Server / Web Tier
Issue       : HTTP service not starting after sudden power shutdown
Service     : adapcctl.sh

 

Issue Faced

After the application server was brought back online, I checked the status of the HTTP service using the EBS admin scripts.

cd $ADMIN_SCRIPTS_HOME
./adapcctl.sh status

This will confirm that the web tier was not running , and users would not be able to access the EBS login page until the HTTP service was restored.

 

Initial Observation

The issue was not related to application tier startup scripts directly. The problem was caused by an OPMN state file that was left behind after the sudden shutdown.

In Oracle EBS 12.2, the Oracle HTTP Server is managed through the web tier components, and OPMN maintains internal state files under the webtier instance directory.

After an abnormal shutdown, these state files can sometimes cause startup issues because the system still holds old or inconsistent runtime information.

 

Solution

To resolve the issue, I went to the OPMN states directory under the webtier instance.

cd /u01/PROD/fs1/FMW_Home/webtier/instances/web_PROD_OHS1/config/OPMN/opmn/states

Then, I listed the files available in this directory.

ls -lrta

The output showed the following files:

-rw-r—–. 1 applprod dba 14  Mar 23 10:58 .locale
-rw-r—–. 1 applprod dba 579 Mar 23 10:59 p1806545876
-rw-r—–. 1 applprod dba 20  Mar 23 11:00 .opmndat

Here, the file p1806545876 was the OPMN state file that needed to be renamed.

Instead of deleting the file directly, I renamed it as a safer approach.

mv p1806545876 p1806545876_bkp

ls -lrt

-rw-r—–. 1 applprod dba 578 Mar 23 10:53 p1806545876_bkp

 

Start the HTTP Service Again

After renaming the problematic OPMN state file, I went back to the EBS admin scripts directory.

cd $ADMIN_SCRIPTS_HOME

./adapcctl.sh start

This time, the HTTP service started successfully.

Validation

After starting the service, I checked the HTTP service status again.

./adapcctl.sh status

The service should now show as running.

You can also validate from the browser by accessing the EBS application URL.

Additionally, it is recommended to monitor the HTTP/OHS logs if the service does not start immediately.

Common locations to check may include:

$LOG_HOME/ora/10.1.3/Apache
$LOG_HOME/ora/10.1.3/opmn
$INST_TOP/logs/ora/10.1.3/Apache
$INST_TOP/logs/ora/10.1.3/opmn

The exact path may vary depending on your EBS 12.2 file system and context configuration.