RZ10 – Character Limitation on Profile Import

I tried to import profile after system refresh a while ago. I just noticed that import profile character entry is limited. As far as I’m involved in SAP Basis consultant, I never experience a problem when importing an instance profile in RZ10. Before I show you where the character limit entry reside, here’s the procedure how I import a profile:

1. Run transaction RZ10.
2. Write down the profile name in Profile
3. Click Import Button (F8)
4. If you receive a warning like ‘First maintain management data‘. Click continue.
5. Fill in the necessary data, Activation in operating system file name and name of reference server for profile parameter check.
   
6. Click on Copy Button
7. Distribute to instance.
8. Save.
9. Profile has been imported.

Now, when you look in the activation in operating system file name field, you will find that it point to physical file which is in Operating System directory. If you are using local directory as in the picture above, this should be no issue on character limit. But if you are using shared directory which has long character name, like for example : \\saphostr3app1\sapmnt\DEV\SYS\profile\START_DVEBMGS00_SAPHOSTR3APP1, you will see as below picture:

When you save this setting you will find your profile file name become START_DVEBMGS00_SAPHO . The character limit is likely 60 character allowed. So just you aware about this limit, you can manage the profile name to a better looks.

Trace SAP Web based Application

When working on SAP Web based application, there will be an issue about internet or network connection and all the things related to web. Applications here like Portal application, BEx Web, ITS connection, Web dynpro, BSP , etc. Sometimes we need to trace those application which can be used to analyze the problem. I’m using HTTP Watch software to trace the loaded scripts and some components. It’s very useful to trace the details of component that been called by SAP Web based application. For example I use HTTP Watch to sniff External ITS missing script files. It will monitor all HTTP and HTTPS based application. You can download the software here. They provide a free software also.  Enjoy!

Transport Control Program TP Error 0208

Problem : Transport control program tp ended with error code 0208. Error shows that transport profile is missing.

Action Steps :

1. Check whether transport profile which is located in Transport Domain exists or not. The link should be like this : //<domain_host>/sapmnt/trans/bin/TP_DOMAIN_<SID>.PFL .
2. Check the TMS configuration like RFC Destination configuration, transport domain setting.
3. Check network connection between localhost and transport domain host.
4. Try to run tp command from OS level. ex : tp showbuffer <SID> pf=<transport_profile>

Apparently, that those check action have nothing wrong. They all are exist and working fine. But still the STMS failed to run any tp programs.

Solution: Check the user SAPservice<SID> in the Windows User Management or in the Domain Controller. It might be locked or expired. In my case, I found that this service user’s password is expired. This makes sense that SAP application cannot call tp program. It use SAPservice<SID> user to run tp program in Operating System.

RSBOLAP018: Java system error

Problem: When you running a Process Chain in BW system, you saw an error RSBOLAP018: Java system error, as shown below.

Investigate: Check log as shown in the error description. You will find this line on files defaultTrace.xx.log :

ERROR RSBOLAP (018): Java system error: An Unknown Error Occured During Portal Communication
MSGV1: An Unknown Error Occured During Portal Communication#

How to find log location is described in my previous post .

Solution : Check your Portal connection from Java to ABAP, ABAP to Java. If the connection is configured properly, so you can retry to execute a process chain again. Why? This issue is explained more detail in SAP  Notes 916090 – RFC error messages in Web Application Designer 2004s. For short description is the process chain needs an active and valid connection during most of the operation. So this could be caused by a network glitch or broken connection which impact on Portal connection error.

Important Log Files for Portal Application

If you run a Portal application based on SAP Web AS, you may find an error on your running process. Mostly the errors are logged into a log files of Java server process. A Portal application usually identified by prefix http://<hostname>:<port>/irj/… Here’s the important log files location :

1. <drive>:\usr\sap\<SID>\DVEBMGS<inst_no>\j2ee\
   cluster\serverX\log\defaultTrace.xx.trc
2. <drive>:\usr\sap\<SID>\DVEBMGS<inst_no>\j2ee\
   cluster\serverX\log\applications.x.log

X: server process number, start from 0,1,2 .. X
xx:default trace number which will be overwritten between 0-19 by default. the most updated can be displayed with the latest timestamp in OS.

The thing is not all people can read what is inside the log files since this is generated roughly from Java process, but those texts are very helpful to predict the portal application. The logs also display some Java class which is used by application. for example :

#1.5#001185E9BA0D0090000031F90000524C00046B19DB1815A0#1243658700982#com.sapportals.wcm.repository.manager.cm.persistence.BasicPersistenceDB#sap.com/irj#com.sapportals.wcm.repository.manager.cm.persistence.BasicPersistenceDB#J2EE_GUEST#0#SAP J2EE Engine JTA Transaction : [0384ffffffd20071ffffffb2]#n/a##0ff0c3b04a1111deb8c3001185e9ba0d#Thread[ThreadPool.Worker1,5,SAPEngine_Application_Thread[impl:3]_Group]##0#0#Error##Plain###Exception occured during garbage collection: com.sapportals.wcm.repository.manager.cm.persistence.PersistenceException: SQLException: Cannot change transaction isolation during distributed transaction and when the connection is shared. DataSource name: “SAPSR3DB”, current isolation level is “TRANSACTION_READ_UNCOMMITTED”, new isolation level is “TRANSACTION_READ_COMMITTED”.#

Yes, of course you need to practice how to solve the issue. There will be a hundreds or even a thousands of variation of the error. Well , prepare your time to learn then

DB02 statistics blank

If I’m not mistaken, display for DB02 is changed from SAP Netweaver 7.0  SP12 into a kind of DB Cockpit which also can be accessed from tx code ST04, DB13, DB14, and some DB transactions. After a Support Package upgrade or refresh sometimes it’s not displayed properly or blank as displayed below :

In my case, to make this transaction displayed correctly, I implement SAP Notes 868063 – Database space statistics monitor for Oracle. Here’s the steps :

1. Run report RSORACUP.
2. Fill input parameter :
   CON_NAME : DEFAULT (for local system), or one of database connection as in F4 help
3. Execute.

ORA-20000:Index or partition of such index is in unusable state

Problem: In SAP BW system when you run update statistics you find an error below,

BR0301E SQL error -20000 at location stats_tab_collect-20, SQL statement:
‘BEGIN DBMS_STATS.GATHER_TABLE_STATS (OWNNAME => ‘”SAPBWP”‘, TABNAME => ‘”/BIC/B0001595000″‘, ESTIMATE_PERCENT
ORA-20000: index “SAPBWP”.”/BIC/B0001595000KE”  or partition of such index is in unusable state
ORA-06512: at “SYS.DBMS_STATS”, line 13267
ORA-06512: at “SYS.DBMS_STATS”, line 13287
ORA-06512: at line 1
BR0886E Checking/collecting statistics failed for table SAPBWP./BIC/B0001595000

Solution : A secondary index for access optimization when reading from a DataSource must be created locally or in partitions for ‘ORACLE’ database systems if tables are partitioned. Follow the procedure belo to solve this :

1. Execute program RSAR_PSA_PARTITION_CHECK from SE38.
   
2. Execute the ‘RSAR_PSA_PARTITION_CHECK’ program in repair mode ( REPAIR = ‘X’ ) if you are prompted to do so in the second row of the log. We recommend that you execute the ‘RSAR_PSA_PARTITION_CHECK’ program in the background.
   
3. Check again without ticking Repair box.
4. If you see display below then it works, otherwise please consult with your BW consultant.
   
5. Done.

Analysing Memory Using sappfpar command in SAP

SAP provides a tools to analyse memory consumption for Web AS ABAP called sappfpar. With this command you can estimate a memory consumption in a detail based on parameter setting in instance profile. Command will be like this :

>sappfpar check pf=<Instance Profile>

and the example result:

From the picture above you can see a lot of parameter on the system which has been set up in the profile. That means you need around 30GB to run this SAP system, so that you can adjust which parameter should be changed. For example : zcsa/table_buffer_area is a table buffer which is set to 110000000 or 104.9 MB. To change tose memory parameters you can go to tcode RZ10 or RZ11 to display.

Reset Password itsadmin in ITS 6.20

Due to incorrect logon or forgot the password user itsadmin in SAP ITS 6.20 is locked. Incorrect password can be occured after upgrading ITS or just after installation. Here’s the procedure how to reset the password:

1. Make a backup of ItsRegistryADM.xml in configuration folder, so you can revert it back if needed.
2. Search “itsadmin” in ItsRegistryADM.xml file.
3. Password value would be like “<value name=”password” type=”text”>des26(7930d45240f)</value>” .
4. Remove the value “des26(7930d45240f)” so that the format would be like this :
   …
   <key name=”Admin”>
   <key name=”Users”>
   <key name=”itsadmin”>
   <value name=”password” type=”text”></value>
   </key>
   </key>
   </key>
   <key name=”Customization”>
   <key name=”Debug”>
   …
5. Save the file and Restart ADM instance.
6. When you first login again, leave the password empty.
7. Change the password in Administration->ChangePassword, there will be a prompted window to fill and confirm the new password.
8. Save and you can logon with new password.
9. Done.

Notes in updating clustered SAP system Kernel

This is a little note when you upgrading the SAP Kernel in clustered system on Windows OS: Dont forget to copy the Kernel folder /usr/sap/<SID>/<instance>/SYS/exe/run to C:/Windows/sapcluster folder. I got an error when I didn’t copy the Kernel folder to cluster directory. And this has to be done in both cluster nodes which I think it should be copied automatically if the cluster system is smart, but somehow it doesn’t work that way when I try to failed over to another cluster. SAP system in SAP MMC will be hanging and it seems dispatcher is on queue trying to connect to mesage server. And after some times it will go down. nice experience ..