Wednesday, September 5, 2012

All About Oracle SOA Suite

Type of BPEL process State 
state =1 --Open and Running
state =2 --Open and Suspended
state =3 --Open and Faulted
state =4 --Closed and (Pending or Cancel)
state =5 --Closed and Completed
state =6 --Closed and Faulted
state =7 --Closed and Cancelled
state =8 --Closed and Aborted
state =9 --Closed and Stale
state =0 --Initiated
--------------------------------------------

How to configure Oracle Application server nodes in cluster.
Change the config topology in opmn.xml file of both nodes. both servers  (nodes) should be on same network.
<opmn>
 <notification-server>
  <port local="6101" remote="6202" request="6004"/>
  <ssl ... />
  <topology>
   <nodes list="node1-sun:6201,node2-sun:6202"/>
  </topology>
 </notification-server>
 ...
</opmn>


Reference: http://download.oracle.com/docs/cd/B25221_05/web.1013/b14432/topology.htm#CHDCFFDB
 
BPEL Process going into Active "Off " State / Inactive /Dead state automatically.
BPEL process with DB Adapter for Polling operation.
    Deployed successfully but upon deployment goes into "OFF" process state by default.
    Changing the Process State to ON --> Apply , doesnot affects or changes back to OFF state automatically.
 
 Cause: 
Polling process is unable to connect to DB, in other word DB Adapter is unable to read from the table or is unable to send the data to the BPEL.

Solution Steps :
1) Check BPEL Connection pools and Datasource by Testing through EM Console. there can be chance of user locked etc.
2) Check DB Adapter and its required resource (data source,Conn pool) 
          Go to EM -> Click BPEL OC4J ->Click Applications -> from drop down choose "Resource Adapter" then click Database Adapter and conifrm db adapter which is configured in polling process is available here. (you can check db adapter name by opening WSDL of polling process in browser). create the adapters if missing. restart BPEL OC4J.

3) Table structure changes : even though there was a change and then reverted back to original, problem may persist
4) Stale/older version of same process exists  --> search and un-deploy the redundant process


How to enable jsso for WSM (draft)
go to $ORACLE_HOME/owsm/bin/install.properties -> update the parameter 'install.sso.support' to true.
deploy owsm applications by -> ./wsmadmin deploy all
supply oc4jadmin password during deployment.
start jsso application from EM for particular OC4J where WSM components are deployed.
Restart ccore, restart wsm by -> ./wsmadmin stop/start
Restart OC4J.

How to change oc4jadmin password (draft)
you have to change in all OC4J containers and BPEL deployment proprty file ant.orabpel.properties)

$ORACLE_HOME/j2ee/home/config/system-jazn-data.xml
<name>oc4jadmin</name>
<display-name>OC4J Administrator</display-name>
<guid>!PWD4admin</guid>

Apache or HTTP_Server is not going down/coming up (draft)
Reason - If you have deleted pid file from $ORACLE_HOME/Apache/Apache/logs/httpd.pid location you will not be able to shutdown apache (Oracle HTTP_SERVER).

Solution - get the pid of httpd process and create a file on above location and open and paste the pid number and save.
ps -ef|grep httpd  --command to get PID for process.

Now try to bring down/Up. it should work. if still not able to bring down then kill the processs manually.

How to clone an application server instance (draft)
perl clone.pl  ORACLE_HOME=/u01/app/uat ORACLE_HOME_NAME=OUIHome1 -instance=SOA1 -db_vendor oracle -db_user system -db_password pwd123 -db_string testdb01:1250 -db_sid testdb -oc4jadmin_old_password oc4j124 -oc4jadmin_new_password oc4j124

How to remove Oracle_HOME (draft)

./runInstaller -deinstall -silent ORACLE_HOME=<LOCATION_OF_ORACLE_HOME> "REMOVE_
HOMES={<LOCATION_OF_ORACLE_HOME_TO_BE_REMOVED>}"


Error: OC4J not coming up after enabaling jmxremote

testap01::soauser>opmnctl startproc process-type=OC4J_WORK
opmnctl: starting opmn managed processes...
================================================================================
opmn id=testap01:6202
    0 of 1 processes started.

ias-instance id=testap.testap01.com
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
--------------------------------------------------------------------------------
ias-component/process-type/process-set:
    NRE/OC4J_Work/default_group/

Error
--> Process (index=1,uid=542644844,pid=28058)
    failed to start a managed process after the maximum retry limit
    Log:
    $ORACLE_HOME/opmn/logs/Work~OC4J~default_group~1.log

testap01::soauser>less $ORACLE_HOME/opmn/logs/Work~OC4J~default_group~1.log

12/10/15 08:50:50 Start process
--------
Error: Password file not found: $ORACLE_HOME/jdk/jre/lib/management/jmxremote.password

Solution:

Confirm/modify  you have below setting in opmn.xml.
-Dcom.sun.management.jmxremote.authenticate=false -Dcom.sun.management.jmxremote.ssl=false -Dcom.sun.management.jmxremote=true -Dcom.sun.management.jmxremote.port=18010

Follow below steps
1. Browse the JRockit/JDK management folder:
$ORACLE_HOME/jdk/jre/lib/management/
2. Rename the file:
jmxremote.password.template
to:
jmxremote.password
3. Remove the write protection from the jmxremote.password file. Context-click on the file and uncheck the Attribute Read-
only. Click on the OK button to close the dialog.
4. Open the jmxremote.password file with your favourite editor and scroll down to the bottom of the document
5. Remove the # in front of:
# monitorRole QED
# controlRole R&D
6.  save and close the document.
Now try starting the OC4J container.



No comments:

Post a Comment