SPUM4 in 4.6C MDMP using CU&UC to upgrade

[Former Member]

Hi Experts,

upgrading from 4.6C MDMP to ECC 6.0 unicode I have ran SPUM4 in 4.6C from scan1 (CC) through scan7 (INDX Repair).

Now I would like to know what is the correct sequence and what is the impact if I run "Update Worklist" now.

I know I will get all tables which have been changed / added since CC check.

If I reset those tables and run CC again will I lose vocabulary entries which have been added by scan2 through scan4 for these tables?

If that is the case I guess I have to run all the scans for these tables again and if new words are found they will be added to the vocabulary as usual. Scanning big cluster tables like CDCLS (even if split) will take quite some time which is my concern.

Another thing: until when am I able to run SPUM4 Update Worklist? Until I start Online Upgrade? Or during online part right before offline part?

Thanks for your answers!

Regards,

Chris

[nils_buerckel]

Hi Chris,

You can execute "update worklist" multiple times.

I would recommend to do the "update worklist" step at least right after the "hard freeze period" has begun. This should leave you some time for rescannings of tables. However in most cases I do not expect that the very large tables (like CDCLS) will be changed in that way that they have to be rescanned. In case you find such a table, then please examine the change carefully and analyze, whether a rescan is necessary.

Then, if there are (known) emergency transports - then I would also recommend to execute the "update worklist" step.

In some cases it might make sense to repeat this procedure (update worklist) periodically during the time when the main spum4 scans have finished and upgrade downtime begins.

Finally, to be sure, you should execute "update worklist" right before the downtime, but due to the hard freeze, there should be no entries there (if there are entries, somebody violated the "hard freeze rules"). However, this would be rather a last check.

Best regards,

Nils Buerckel

SAP AG

[Former Member]

Thanks a lot.

[Former Member]

Hi Chris,

I will be doing the Upgrade & unicode conversion for one my customer having R/3 4.6C in some days. I gone through the CU&UC document from the service site. But i find it confusing for MDMP system. Request you to help me to know the sequence of activities to be done to convert MDMP system as you already done it.

Your reply will be help me Immensely. Request you to reply.

Regards

[nils_buerckel]

Hi,

I think you have to be more specific here, otherwise feedback will be low.

I am copying the MDMP tasks here from the Guide:

2.3 MDMP Preconversion

2.3.1 Manually Assign the Code Page for an Entire Table

2.3.2 Creating the Vocabulary

2.3.2.1 Scan 2: Tables without Language Information

2.3.2.2 Scan 3: Tables with Ambiguous Language Information

2.3.2.3 Edit Invalid Language Keys

2.3.2.4 INDX-type Tables

2.3.2.5 Scan 4: INDX Analysis

2.3.3 Update Database Statistics

2.3.4 Maintaining the Vocabulary

2.3.4.1 Assignment Options

2.3.4.2 Vocabulary Transfer

2.3.4.3 Scan 5: Tables with Language Information

2.3.4.4 Hint Management

2.3.4.5 Manually Assign Language

2.3.5 Scan 6: Reprocess

2.3.5.1 Reprocess Log

2.3.6 Scan 7: INDX Repair

2.3.6.1 INDX Log

2.3.7 Resolving Collisions

These are basically the tasks you have to do for all systems (with reusage of results from former conversions).

Please specify, which step is confusing.

Maybe the attachment of SAP note 1037613 helps you to understand the general concept of an MDMP conversion.

Best regards,

Nils Buerckel

SAP AG