Monday, May 28, 2012

Bouncing the Oracle Applications in Multi-node environment Configured with LOAD BALANCING,SHARED_APPL TOP and RAC setup


Hi DBAs,
Below are the steps to Bounce the Oracle Applications in SHARED_APPL TOP environment:

Consider we have the application  APPS_TESTAPPS.

Note:This is done in Oracle EBS R12 Environment where we have $ADMIN_SCRIPTS_HOME a common place where you find all the start and stop scripts .

Step1)Stopping the apps listener on node1 and node2
==========================================================
 
a) stop apps listener on node1:

=>APPS_TESTAPPS
node1/product/app/TESTAPPS/inst/apps/TESTAPPS1_node1/admin/scripts >./adalnctl.sh stop

adalnctl.sh version 120.3

Shutting down listener process APPS_TESTAPPS.

adalnctl.sh: exiting with status 0


adalnctl.sh: check the logfile /product/app/TESTAPPS/inst/apps/TESTAPPS1_node1/logs/appl/admin/log/adalnctl.txt for more information ...

b)Stop apps listener on node2:

node2 /product/app/TESTAPPS/inst/apps/TESTAPPS1_node2/admin/scripts >./adalnctl.sh stop

adalnctl.sh version 120.3

Shutting down listener process APPS_TESTAPPS.

adalnctl.sh: exiting with status 0


adalnctl.sh: check the logfile /product/app/TESTAPPS/inst/apps/TESTAPPS1_node2/logs/appl/admin/log/adalnctl.txt for more information .


Step 2:We stop all the Middle tier or Oracle applications service on node1....
==============================================================================

node1/product/app/TESTAPPS/inst/apps/TESTAPPS1_node1/admin/scripts >./adstpall.sh

Step 3:We stop all the Middle tier or Oracle applications service on node2....
==========================================================================

node2/product/app/TESTAPPS/inst/apps/TESTAPPS2_node2/admin/scripts >./adstpall.sh


step 4): Kill all processes by finding the TESTAPPS processes:
==============================================================

node 1:
ps -ef|grep TESTAPPS

kill -9 pid.

node 2:
ps-ef|grep TESTAPPS

kill -9 pid


Step 5: Start apps listener on node1 and node2
================================================


Node 1:
 node1/product/app/TESTAPPS/inst/apps/TESTAPPS1_node1/admin/scripts >./adalnctl.sh start

adalnctl.sh version 120.3

Checking for FNDFS executable.
Starting listener process APPS_TESTAPPS.

adalnctl.sh: exiting with status 0


adalnctl.sh: check the logfile /product/app/TESTAPPS/inst/apps/TESTAPPS1_node1/logs/appl/admin/log/adalnctl.txt for more information ...

Node 2:

node2 /product/app/TESTAPPS/inst/apps/TESTAPPS1_node2/admin/scripts >./adalnctl.sh start

adalnctl.sh version 120.3

Checking for FNDFS executable.
Starting listener process APPS_TESTAPPS.

adalnctl.sh: exiting with status 0


adalnctl.sh: check the logfile /product/app/TESTAPPS/inst/apps/TESTAPPS1_node2/logs/appl/admin/log/adalnctl.txt for more information ...

If we have started the concurrent managers from the primary node,than make sure they are up on both the nodes and rename the adcmctl.sh script on both the primary and secondary nodes as adcmctl_bk.sh.Now start the middle Tier applications on both the nodes.


Step 6:Start all the middle Tier or Oracle Applications from node1.....
======================================================================================



node1/product/app/TESTAPPS/inst/apps/TESTAPPS1_node1/admin/scripts >./adstrall.sh

Step 7:Start all the middle Tier or Oracle Applications from node2.....
======================================================================================



node2/product/app/TESTAPPS/inst/apps/TESTAPPS2_node1/admin/scripts >./adstrall.sh

Step 8:Verify the applications of TESTAPPS started on both the nodes.
=======================================================


Node 1:

ps -ef|grep TESTAPPS

Node 2:

ps -ef|grep TESTAPPS

Hope it helps...

Enjoy Apps DBA learning....



Best regards,

Rafi.









Starting and Stopping Concurrent Managers in PCP Environment

     Hi DBAs,
Below are the steps to Bounce the Concurrent Managers in the PCP(Parallel Concurrent Processing) Environment.Important point to note is we can start all the concurrent Managers running on different nodes from  the node which is primary node and not required to start on both nodes to avoid the situation of failover.In below environment node1 is the primary node of PCP setup.

Note:How to check the primary node in PCP setup given in below link:
http://www.blogger.com/blogger.g?blogID=8370545678641066072#editor/target=post;postID=6995260641016133607

step1)Stopping the apps listener on node1(Node1):
===========================================================



a) stop apps listener
=>APPS_TESTAPPS
Node1 /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node1/admin/scripts >./adalnctl.sh stop

adalnctl.sh version 120.3

Shutting down listener process APPS_TESTAPPS.

adalnctl.sh: exiting with status 0


adalnctl.sh: check the logfile /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node1/logs/appl/admin/log/adalnctl.txt for more information ...


b)Stop apps listener on node2(Node2)

Node2 /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node2/admin/scripts >./adalnctl.sh stop

adalnctl.sh version 120.3

Shutting down listener process APPS_TESTAPPS.

adalnctl.sh: exiting with status 0


adalnctl.sh: check the logfile /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node2/logs/appl/admin/log/adalnctl.txt for more information ...

step 2)Stop the apps concurrent manager
========================================

adcmctl.sh stop

 Node1 /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node1/admin/scripts >./adcmctl.sh stop

You are running adcmctl.sh version 120.17.12010000.5


Enter the APPS username : APPS

Enter the APPS password :
Shutting down concurrent managers for TESTAPPS_CM2 ...
ORACLE Password:
Submitted request 2532852 for CONCURRENT FND SHUTDOWN


adcmctl.sh: exiting with status 0


adcmctl.sh: check the logfile /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node1/logs/appl/admin/log/adcmctl.txt for more information ...


step 3): Kill all processes by finding the concurrent processes:
=====================================================================


ps -ef|grep FNDLIBR|grep appstest

kill -9 pid...


Step 4: Start apps listener on node1 and node2
=====================================================


Node 1:
-------------
 Node1 /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node1/admin/scripts >./adalnctl.sh start

adalnctl.sh version 120.3

Checking for FNDFS executable.
Starting listener process APPS_TESTAPPS.

adalnctl.sh: exiting with status 0


adalnctl.sh: check the logfile /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node1/logs/appl/admin/log/adalnctl.txt for more information ...

Node 2:
--------

Node2 /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node2/admin/scripts >./adalnctl.sh start

adalnctl.sh version 120.3

Checking for FNDFS executable.
Starting listener process APPS_TESTAPPS.

adalnctl.sh: exiting with status 0


adalnctl.sh: check the logfile /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node2/logs/appl/admin/log/adalnctl.txt for more information ...


Step 5:Start the concurrent manager on any one node....
=========================================================



Node1 /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node1/admin/scripts >./adcmctl.sh start

You are running adcmctl.sh version 120.17.12010000.5


Enter the APPS username : APPS

Enter the APPS password :
Starting concurrent manager for TESTAPPS_CM2 ...
Starting TESTAPPS_0216@TESTAPPS_CM2 Internal Concurrent Manager
Default printer is noprint

adcmctl.sh: exiting with status 0


adcmctl.sh: check the logfile /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node1/logs/appl/admin/log/adcmctl.txt for more information ...

 Node1 /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node1/admin/scripts >ps -eaf|grep FNDLIBR
  appstest 18179 18170   0 04:46:03 pts/12      0:00 FNDLIBR FND CPMGR FNDCPMBR sysmgr="" sleep=30 pmon=4 quesiz=1 diag=N logfile=/p
  appstest 18280 16419   0 04:46:25 pts/12      0:00 grep FNDLIBR
 Node1 /product/app/TESTAPPS/inst/apps/TESTAPPS1_Node1/admin/scripts >ps -eaf|grep FNDSM
  applqb 15336     1   0   Feb 12 ?           0:04 FNDSM
  appstest 18556 16419   0 04:46:50 pts/12      0:00 grep FNDSM
  appstest 18554 18179   0 04:46:50 pts/12      0:00 tnsping FNDSM_Node2_TESTAPPS

 Note:If any services is already started and if you try to start again,
we will get message status code 2 message and if it start successfully in first attempt
 we will see status code 0,there is no harm in trying to start the services which are already running but please verify the services and Check EBS home page by login as 'SYSADMIN' and check all the managers are up & running and Concurrent Processing Server is up & runing on both the node.

Step 6:Verify on both the nodes:
==========================

Node 1:
$ps -ef|grep FNDLIBR|grep appstest



Node 2:
$ps -ef|grep FNDLIBR|grep appstest

 If it shows the processes related with appstest,than Concurrent Managers are running on both the nodes in the PCP setup.


Hope it helps...

Enjoy Apps DBA learning...

Best regards,
Rafi.