J2ee process restarts

[Former Member]

Hello,

On our portal instance here, J2ee process restarts automatically and randomly. It has been working fine for sometime. But now, I am not sure for what reason, it restarts by itself. I find huge entries like this in our developer trace file of J2ee process..

[Unloading class jsp._sapportalsjsp_NavigationForms]

[Unloading class sun.reflect.GeneratedMethodAccessor250]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor232]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor179]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor299]

[Unloading class sun.reflect.GeneratedMethodAccessor313]

[Unloading class sun.reflect.GeneratedMethodAccessor306]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor107]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor6]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor393]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor14]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor255]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor104]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor451]

These are the JVM parameters for our server process..

-Djava.security.policy=./java.policy

-Djava.security.egd=file:/dev/urandom

-Dorg.omg.CORBA.ORBClass=com.sap.engine.system.ORBProxy

-Dorg.omg.CORBA.ORBSingletonClass=com.sap.engine.system.ORBSingletonProxy

-Djavax.rmi.CORBA.PortableRemoteObjectClass=com.sap.engine.system.PortableRemoteObjectProxy

-Djco.jarm=1

-XX:MaxPermSize=256M

-XX:PermSize=256M

-Xms1024M

-XX:NewSize=171M

-XX:MaxNewSize=171M

-XX:+DisableExplicitGC

-verbose:gc

-Xloggc:GC.log

-XX:+PrintGCDetails

-XX:+PrintGCTimeStamps

-Djava.awt.headless=true

-Dsun.io.useCanonCaches=false

-XX:SoftRefLRUPolicyMSPerMB=1

-XX:SurvivorRatio=2

-XX:TargetSurvivorRatio=90

-XX:+UseParNewGC

-XX:+HandlePromotionFailure

-Dorg.omg.PortableInterceptor.ORBInitializerClass.com.sap.engine.services.ts.jts.ots.PortableInterceptor.JTSInitializer

-Djava.io.tmpdir=C:\WINDOWS\Temp

-Dcm.tmpdir=C:\WINDOWS\Temp

We are currently on Windows 2003 SP2 version (32 bit version), J2ee 7.0 and MS SQL 2005 version. We have JDK 1.4.2_13 ver installed on our servers..

I tried increasing the max heap size (-Xmx) and initial heap size (-Xms) to a higher value. But I guess for 32 bit machines, it is not allowed beyond 1024 MB.

Please let me know your thoughts on this issue and how to avoid the restart from happening.

Regards

Srinidhi Rayasam

[Former Member]

Hi,

I have faced this problem .

work around done from our side :

1.Make sure your portal server has 8GB RAM for stable working of portal.

2.Stop any code coverage tool,if it is activated.

3.Go to c:/usr/sap/SID/JCxx/j2ee/configtool----configtool.bat.

Go to Instance node,go to server general tab under that.

Change values of -Xmx and -Xms to 2048.

-Newsize values to 320

Permsize values to 512.

These changes should be done under all server nodes.

No changes need to be done under dispatcher node.

Also change bootstrap values to 545.

Save and restart the instance.

It will be stable.

reward points if helpful...........

[Former Member]

I have the same problem, and if this class crashes it reduce the memory usage of the server node from 2887900K to 929362K. We had in past a lot of out of memory errors and increased the memory for the server node from 2GB to 3GB. I think the java node as ha memory leak? Can I find out the involved component? More often the GRMG monitoring means the BI UDC component is not available (/BI_UDC/servlet/GrmgServlet), but I could not find more details about this.

Mon Feb 07 14:44:59 2011

413626.267: [GC 413626.267: [ParNew: 261888K->53710K(261888K), 0.4932280 secs] 2887900K->2756761K(3058432K), 0.4933676 secs]

Mon Feb 07 14:45:00 2011

413627.455: [GC 413627.455: [ParNew: 228302K->228302K(261888K), 0.0000417 secs]413627.455: [Tenured

Mon Feb 07 14:45:01 2011

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor8508]

[Unloading class sun.reflect.GeneratedMethodAccessor3964]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor8842]

[Unloading class sun.reflect.GeneratedMethodAccessor3685]

[Unloading class sun.reflect.GeneratedMethodAccessor3572]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor8817]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor8418]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor8177]

[Unloading class sun.reflect.GeneratedMethodAccessor3314]

...

...

...

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor9209]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor8979]

[Unloading class sun.reflect.GeneratedConstructorAccessor1143]

[Unloading class sun.reflect.GeneratedMethodAccessor3792]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor9164]

[Unloading class sun.reflect.GeneratedConstructorAccessor1186]

[Unloading class sun.reflect.GeneratedSerializationConstructorAccessor9047]

[Unloading class jsp._sapportalsjsp_HeaderiView]

Mon Feb 07 14:45:03 2011

: 2703050K->754770K(2796544K), 2.8378332 secs] 2931353K->754770K(3058432K), 2.8380830 secs]

413631.028: [GC 413631.028: [ParNew: 174592K->20286K(261888K), 0.0478163 secs] 929362K->775057K(3058432K), 0.0479335 secs]

[Former Member]

This was an error with increased memory values:

Wed Feb 02 19:50:40 2011

163449.060: [GC 163449.060: [ParNew: 230639K->78351K(261888K), 0.2526526 secs] 2928558K->2776270K(3058432K), 0.2527474 secs]

163449.635: [GC 163449.635: [ParNew

Wed Feb 02 19:50:41 2011

Java HotSpot(TM) 64-Bit Server VM warning: increase BUFLEN in ostream.cpp -- output truncated

[Thr 3232] JLaunchIAbortJava: abort hook is called

[Thr 3232] **********************************************************************

• ERROR => The Java VM aborted unexpectedly.

• Please see SAP Note 943602 , section 'Java VM crashes'

• for additional information and trouble shooting.

**********************************************************************

[Thr 3232] JLaunchCloseProgram: good bye (exitcode = -2)

This Error was more often before with 2GB memory: out of memory

[Thr 7408] JLaunchIExitJava: exit hook is called (rc = 666)

[Thr 7408] **********************************************************************

• ERROR => The Java VM terminated with a non-zero exit code.

• Please see SAP Note 943602 , section 'J2EE Engine exit codes'

• for additional information and trouble shooting.

**********************************************************************

[Thr 7408] JLaunchCloseProgram: good bye (exitcode = 666)

[Former Member]

Hi

What is your swap space.

Thanks,

Tanuj

[Former Member]

Hi,

My current swap size is 6179 MB and the recommended value is 10240 MB.

[Former Member]

Hi,

Set the Java VM Accordingly.

723909 Java VM settings for J2EE 6.30/6.40/7.0

Thanks,

Tanuj

[Former Member]

Hello Tanuj,

I looked into these SAP notes, and as per recommendations, our current JVM parameters config meets most of the recommended values I guess.

Does this have to do with swap size in any case? Since we got a warning message during our installation saying that our swap size was below the recommended settings?