After Homo sys copy (detach/attach), dispatcher, server0 not starts

[Former Member]

ALL,

I am facing a issue where my (XI 7.0 on win2k3-MSSQL 2005) abap stack is up......but dispatcher & server0 is stopping with exit code -337.

Below are the ending extarct of trace files.

std_dispatcher.out

..
...
....
CompilerOracle: exclude com/sap/engine/services/webservices/jaxrpc/encoding/InstanceBuilder readElement
CompilerOracle: exclude com/sap/engine/services/webservices/jaxrpc/encoding/InstanceBuilder readSequence
CompilerOracle: exclude com/sap/engine/services/webservices/jaxrpc/encoding/TypeMappingImpl initializeRelations
CompilerOracle: exclude com/sap/engine/services/webservices/jaxrpc/encoding/GeneratedComplexType _loadInto
SAP J2EE Engine Version 7.00   PatchLevel 108458.44 is starting...
Loading: LogManager ... 344 ms. 
Loading: PoolManager ... 0 ms. 
Loading: ThreadManager ... 31 ms. 
Loading: IpVerificationManager ... 16 ms. 
Loading: ConnectionsManipulator ... 15 ms. 
Loading: ClassLoaderManager ... 0 ms. 
Loading: ClusterManager ... [Framework -> criticalShutdown] Exiting Listener Loop. This requires a restart of the node. Possible reason is an interrupted reconnect session to the message server.
Aug 24, 2008 5:11:33 PM             com.sap.engine.core.Framework [SAP J2EE Engine|MS Socket Listener] Fatal: Critical shutdown was invoked. Reason is: Exiting Listener Loop. This requires a restart of the node. Possible reason is an interrupted reconnect session to the message server.
Aug 24, 2008 5:11:33 PM  ...anagerImpl.init(java.util.Properties) [Thread[Thread-1,5,main]] Fatal: Cannot attach to the Message Server. cluster Id is not unique.
Loading: ClusterManager returned false! 
Kernel not loaded. System halted.

std_server0.out

CompilerOracle: exclude com/sap/engine/services/webservices/jaxrpc/encoding/TypeMappingImpl initializeRelations
CompilerOracle: exclude com/sap/engine/services/webservices/jaxrpc/encoding/GeneratedComplexType _loadInto
SAP J2EE Engine Version 7.00   PatchLevel 108458.44 is starting...
Loading: LogManager ... 360 ms. 
Loading: PoolManager ... 0 ms. 
Loading: ApplicationThreadManager ... 46 ms. 
Loading: ThreadManager ... 16 ms. 
Loading: IpVerificationManager ... 0 ms. 
Loading: ClassLoaderManager ... 16 ms. 
Loading: ClusterManager ... [Framework -> criticalShutdown] Exiting Listener Loop. This requires a restart of the node. Possible reason is an interrupted reconnect session to the message server.
Aug 24, 2008 5:11:33 PM             com.sap.engine.core.Framework [SAP J2EE Engine|MS Socket Listener] Fatal: Critical shutdown was invoked. Reason is: Exiting Listener Loop. This requires a restart of the node. Possible reason is an interrupted reconnect session to the message server.
Aug 24, 2008 5:11:33 PM  ...anagerImpl.init(java.util.Properties) [Thread[Thread-1,5,main]] Fatal: Cannot attach to the Message Server. cluster Id is not unique.
Loading: ClusterManager returned false! 
Kernel not loaded. System halted.

dev_jcontrol

[Thr 3132] JControlICheckProcessList: process server0 started (PID:4668)
[Thr 3132] Sun Aug 24 17:11:37 2008
[Thr 3132] JControlICheckProcessList: process dispatcher (pid:5380) died (RUN-FLAG)
[Thr 3132] JControlIResetProcess: reset process dispatcher
[Thr 3132] JControlIResetProcess: [dispatcher] not running -> increase error count (4)
[Thr 3132] JControlICheckProcessList: running flight recorder:
	C:\j2sdk1.4.2_12\bin\java -classpath ../j2ee/cluster/bootstrap/sap.comtcbloffline_launcherimpl.jar com.sap.engine.offline.OfflineToolStart com.sap.engine.flightrecorder.core.Collector ../j2ee/cluster/bootstrap -node ID18973500 1219612292 -bz D:\usr\sap\RDC\SYS\global
[Thr 3132] JControlICheckProcessList: process server0 (pid:4668) died (RUN-FLAG)
[Thr 3132] JControlIResetProcess: reset process server0
[Thr 3132] JControlIResetProcess: [server0] not running -> increase error count (4)
[Thr 3132] JControlICheckProcessList: running flight recorder:
	C:\j2sdk1.4.2_12\bin\java -classpath ../j2ee/cluster/bootstrap/sap.comtcbloffline_launcherimpl.jar com.sap.engine.offline.OfflineToolStart com.sap.engine.flightrecorder.core.Collector ../j2ee/cluster/bootstrap -node ID18973550 1219612292 -bz D:\usr\sap\RDC\SYS\global

dispatcher defaultTrace.0.trc

#
#1.5 #001B7838D5B6000200000004000015040004553B1D89DEC4#1219612293848#com.sap.engine.core.thread.impl6.SingleThread##com.sap.engine.core.thread.impl6.SingleThread.SingleThread.execute().RuntimeException#######SAP J2EE Engine|MS Queue Listener##0#0#Error##Plain###java.lang.NullPointerException#
#1.5 #001B7838D5B6000200000005000015040004553B1D89DF8E#1219612293848#com.sap.engine.core.thread.impl6.SingleThread##com.sap.engine.core.thread.impl6.SingleThread.SingleThread.execute().RuntimeException#######SAP J2EE Engine|MS Queue Listener##0#0#Error##Plain###java.lang.NullPointerException
	at com.sap.engine.core.cluster.impl6.ms.MSListenerQueue.run(MSListenerQueue.java:71)
	at com.sap.engine.frame.core.thread.Task.run(Task.java:64)
	at com.sap.engine.core.thread.impl6.SingleThread.execute(SingleThread.java:78)
	at com.sap.engine.core.thread.impl6.SingleThread.run(SingleThread.java:148)
#
#1.5 #001B7838D5B600000000000E000015040004553B1D8CF29B#1219612294051#com.sap.engine.core.thread.impl6.ThreadManagerImpl##com.sap.engine.core.thread.impl6.ThreadManagerImpl#######Thread[Thread-1,5,main]##0#0#Error##Plain###Unexpected thread activity after interrupt() is executed in shutdown of SAPEngine_System_Thread[impl:6]_ThreadManager:
Thread[SAP J2EE Engine|MS Socket Listener]
Task: com.sap.engine.core.thread.impl6.ActionObject - Processing Task [classname: com.sap.engine.core.cluster.impl6.ms.MSListener | toString: com.sap.engine.core.cluster.impl6.ms.MSListener@76ab2f] with classloader [com.sap.engine.boot.CoreClassLoader@1503a3]#

Please help.....

Waiting for a reply asap.....

[markus_doehr2]

Did you run the SQL-Server STM tools after the attach?

Markus

[Former Member]

Markus,

The SQL server STM tools cannot be used for double (ABAP+JAVA) stack system.

Correct me if i am wrong?

Regards

Abb

[markus_doehr2]

How did you do the copy? Using sapinst? If this is a combined ABAP + Java instance, sapinst must be used. The reason is:

Much configuration data of the Java instance is not stored in the database but locally on the filesystem and in the database. If you copy an instance manually you would need to go through all the properties files in configtool and adapt everything manually (hostnames, user/passwords...). This is done by using sapinst. If you didn´t use sapinst, you need to start again doing the copy with that.

(http://service.sap.com/systemcopy)

And for the original question: Check note "151603 - Copying an SQL Server database", read section 8, first sentence.

Markus

[Former Member]

Markus,

My source system (Prdouction) downtime is not possible as of now.... so i am using the backup taken by OS team which is totally flat files backup.

I manually copied all the datafiles to the target system and attcahed it to sql instances and did some security realted work by which my config tool on the target system started connecting to the new database.

After doing some more troubleshooting I started to see the j2ee process tables in the SAP MMC and now the server0 and dispatcher doesnt come up at all.

So i am stuck here.

Should I copy the usr folder of the source system and replace it with target system.... so should i copy specfic files from the source usr folder.

Note: my target SID and source SID are different but instance numbers are the same.

I feel their is somewhere mismatch between the OLD cluster IDXXXXXX and the new Cluster IDXXXXX. But I cant figure out where it is ?

Please help.

Thanks in advance.

[markus_doehr2]

I'm sorry - but this doesn't work.

If you restore the flat files under \usr\sap\<SID>\DVEBMGS<NO>\j2ee all the .properties files and the configurations stored in the database contain references to the source system. The process which of those files must be changed and in which database table of the Java schema the references are "wrong" is not documented - and it depends on what you have installed on the engine (Business Packages etc.)

The only way to do Java systemcopies is using sapinst, this is unfortunately the only supported one. Sapinst "knows" what needs to be changed in all the .properties and java database schema in order to bring the engine up again.

Markus

[Former Member]

Hi,

As you said:

The only way to do Java systemcopies is using sapinst, this is unfortunately the only supported one. Sapinst "knows" what needs to be changed in all the .properties and java database schema in order to bring the engine up again.

I am thinking of a sitiuation when there is a disaster and the source system is not available to perform an export via sapinst.... so in that case what should be done?

Let me know ur thoughts????

Regards

[markus_doehr2]

You can, of course, use the flat file backup you´ve done but you must keep

- the hostname/full qualified domain name and

- the SID

of the instance the same. Then the restore of the instance will work in case of a DR.

Markus

[Former Member]

Markus,

If I follow what u say..... then I face the problem mentioned in the below thread(Link).

[;

Since i didnt got much help from the above thread.... I thought of this option (detach/attach), but the current option also seems to be not working....

U had also put in ur thoughts in the above thread.

Please advice.

[markus_doehr2]

I feel your pain

Why not install the DR server with the virtual hostname from the cluster?

Markus

[Former Member]

Thanks Markus for understanding my situation

The question is How do I start it?

Secondly, in HA setup there are 2 Cluster id .... one for the CI and one for Dialog instance..... so in the source database I have 2 clusterids and when I copy the source file to target... it will not work .

Since my DR is non HA mode

[markus_doehr2]

> The question is How do I start it?

Rename the machine (computername + hostname) to the virtual name of the cluster?

> Secondly, in HA setup there are 2 Cluster id .... one for the CI and one for Dialog instance..... so in the source database I have 2 clusterids and when I copy the source file to target... it will not work .

I'm not sure if that, what you're trying to accomplish, is possible. You have a distributed instance and you're trying to set up a DR system which is not distributed.

The question is: What are you trying to do? Can't you "just" add the DR server as third standby instance to the cluster?

Markus