Archive

Archive for October, 2013

Job step scheduler showing as down in EM12c

October 25, 2013 Leave a comment

Image


a)Ensure that OS level timezone and the Timezone value set in the Database are in Sync.

Verify it with the following commands:

$date

This gives the timestamp on the Machine with Timezone.

b)Connect to the repository Database and run the following query to check the value set in the Database.

SQL> select dbms_scheduler.stime from dual;

If there is any difference, run the following as SYS user:

SQL> exec DBMS_SCHEDULER.SET_SCHEDULER_ATTRIBUTE('default_timezone','TimeZone');

For example:

SQL> exec DBMS_SCHEDULER.SET_SCHEDULER_ATTRIBUTE('default_timezone','Europe/London');

If there is no difference in the Time Zone values, proceed with the below Action plan.
 c)Connect as SYS user and Set the job_queue_processes to zero.

SQL>alter system set job_queue_processes=0 scope=BOTH;

SQL> commit;

d) Connect as SYSMAN user and execute the following:
 SQL>exec emd_maintenance.remove_em_dbms_jobs;

SQL> commit;

e) Reset the value of the following parameter to 1000 as SYS user:

SQL>alter system set job_queue_processes=1000 scope=BOTH;

f) Submit DBMS_SCHEDULER jobs. (Connect as SYSMAN user and execute the following:)

SQL>exec emd_maintenance.submit_em_dbms_jobs;

SQL>commit;

g)Verify the status of job on the console now

Advertisements
Categories: cloud control1 2c

Changing the heap size of the OMS in EM12c

October 18, 2013 1 comment
      version 12.1.0.3

We had an issue where oracle management server ( OMS) was restarted on a daily basis  during the hourrs of 2 – 6 am , after numerous support updates with Oracle got no further forward , All they could suggest was increase the RAM of the machine as it was set to 8GB and they wanted me to increase it to 12GB

the error message being reported in the logs was

HealthMonitor 16-Oct-2013 03:38:18 OMS Heartbeat Recorder: error: Zero rows updated for this oms
during heartbeat
Critical error err=3 detected in module OMS Heartbeat Recorder:
OMS will be restarted.
A full thread dump will be generated in the log file

/u01/app/oracle/gc_inst1/user_projects/domains/GCDomain/servers/EMGC_OMS1/logs/EMGC_OMS1.out

to help Oracle Support analyse the problem.

Please consult My Oracle Support Note 964469.1 for additional information.

The above error message could be related to a number of things but after speaking
with our system admin he did some investigation and notice that the JVM out of the box
was set to an initial size of 256 MB and a maximum size of 1750 MB which meant it
will have been constantly resizing itself and getting fragmented

So during garbage collection it would start impacting the performance of your Enetrprise manager

backup the startEMServer.sh script, then increase the Java heap size

cd /u01/app/oracle/middleware/oms12c/gc_inst1/user_projects/domains/GCDomain/bin
cp startEMServer.sh startEMServer.sh.bak
vi startEMServer.sh

Look for the first USER_MEM_ARGS reference within the file and change from

if [ "${SERVER_NAME}" != "EMGC_ADMINSERVER" ] ; then
USER_MEM_ARGS="-Xms256m -Xmx1024m -XX:MaxPermSize=612M -XX:-DoEscapeAnalysis -XX:+UseCodeCacheFlushing
-XX:ReservedCodeCacheSize=100M -XX:+UseConcMarkSweepGC -XX:+UseParNewGC -XX:+CMSClassUnloadingEnabled"
if [ "${JAVA_VENDOR}" = "Sun" ] ; then
if [ "${PRODUCTION_MODE}" = "" ] ; then
 USER_MEM_ARGS="-Xms256m -Xmx1024m -XX:MaxPermSize=612M -XX:-DoEscapeAnalysis -XX:+UseCodeCacheFlushing
-XX:ReservedCodeCacheSize=100M -XX:+UseConcMarkSweepGC -XX:+UseParNewGC -XX:+CMSClassUnloadingEnabled -XX:CompileThreshold=8000 -XX:PermSize=128m"

to this , we are fixing the size to 2560M

if [ "${SERVER_NAME}" != "EMGC_ADMINSERVER" ] ; then
USER_MEM_ARGS="-Xms2560m -Xmx2560m -XX:MaxPermSize=612M -XX:-DoEscapeAnalysis -XX:+UseCodeCacheFlushing
-XX:ReservedCodeCacheSize=100M -XX:+UseConcMarkSweepGC -XX:+UseParNewGC -XX:+CMSClassUnloadingEnabled"
if [ "${JAVA_VENDOR}" = "Sun" ] ; then if [ "${PRODUCTION_MODE}" = "" ] ; then
 USER_MEM_ARGS="-Xms2560m -Xmx2560m -XX:MaxPermSize=612M -XX:-DoEscapeAnalysis -XX:+UseCodeCacheFlushing
-XX:ReservedCodeCacheSize=100M -XX:+UseConcMarkSweepGC -XX:+UseParNewGC -XX:+CMSClassUnloadingEnabled -XX:CompileThreshold=8000 -XX:PermSize=128m"

restart your OMS to the changes to take effect:


export OMS_HOME= /u01/app/oracle/middleware12.1.0.3/oms12c/oms
$OMS_HOME/bin/emctl stop oms
$OMS_HOME/bin/emctl start oms

Check that your OMS has started up with the correct memory settings:

grep "JAVA Memory arguments" /u01/app/oracle/middleware/oms12c/gc_inst1/user_projects/domains/GCDomain/servers/EMGC_OMS1/logs/EMGC_OMS1.out

JAVA Memory arguments: -Xms2056m -Xmx2056m -XX:MaxPermSize=1024M -XX:-DoEscapeAnalysis -XX:+UseCodeCacheFlushing -XX:ReservedCodeCacheSize=100M -XX:+UseConcMarkSweepGC -XX:+UseParNewGC -XX:+CMSClassUnloadingEnabled -XX:CompileThreshold=8000 -XX:PermSize=128m#

this vastly improved performance and stopped the daily restarts

Categories: cloud control1 2c

EM12c upgrade from 12.1.0.2 to 12.1.0.3

October 18, 2013 Leave a comment
Categories: cloud control1 2c

Convery Single Instance to RAC Part 2

October 7, 2013 Leave a comment

Convery Single Instance to RAC Part 2

Step by step guide to covert Single Instance to RAC Part 2 by Satisbabu Gunukula

Categories: Oracle Rac

Covert single instance to RAC Part 1

October 7, 2013 Leave a comment

Covert single instance to RAC Part 1

Step by step guide to covert Single Instance to RAC Part 1 by Satisbabu Gunukula

Categories: Oracle Rac

cleaning up my oracle recommendation from Oracle enterprise manager 12c

October 4, 2013 Leave a comment

recommedations

login as sysman into repository database and clean up the recommendations by running

delete from mos_pa_final_recom;
commit;

reset the checksums so we  need do a full re-computation


exec em_intsdk_patch_central.set_patch_recom_flag('N');

commit;
exec em_intsdk_patch_central.set_patch_recom_flag('Y');

commit

Run the Refresh From My Oracle Support  job from Enterprise Manager to Evaluate the recommendation

Categories: cloud control1 2c