Skip to end of metadata
Go to start of metadata

About it

This is a report used to check the consistency of data from cluster tables (old records which were not accessed for some time), it is normally recommended to run before performing some processes such as Unicode Conversion.

Documentation

SAP Notes:

1309728 - SDBI_CLUSTER_CHECK: Usage instructions

1077403 - Cluster table check with SDBI_CLUSTER_CHECK

FAQ

This page is part of:

DB Independent Database Interface Wiki

Tiny link (useful for e-mail):

https://wiki.scn.sap.com/wiki/x/JRKGGg

Can I run SDBI_CLUSTER_CHECK without option "Find duplicates"? Is there any impact?

Yes, it is possible to run the report without any impact. The option "find duplicates" is also called "sort check". The cluster interface relies on the logic records being sorted properly by the rest key, which is checked when the option "find duplicates" is enabled.

How can I know more about the runtime of SDBI_CLUSTER_CHECK?

SAP Note 1309728 describes the runtime.

Is there an alternative/faster option to check the cluster tables other than SDBI_CLUSTER_CHECK?

Yes, It is possible to use the R3check to perform the same check, further information, check SAP Note 89384.

Can I run the SDBI_CLUSTER_CHECK job months in advance?

Yes, you can run the SDBI_CLUSTER_CHECK in advance. The reason to run in advance is to have enough time to fix issues, if necessary.

If I run report SDBI_CLUSTER CHECK in advance and later install support package or new cluster tables are added to the system, do I need to run it again?

No, usually SDBI_CLUSTER_CHECK finds old issues in data that was not touched for years. New corruptions on live data are likely raise errors immediately, thus the probability for new corruptions that nobody notices is low. 

My quality/development system is a copy from production system, can I skip the SDBI_CLUSTER_CHECK?

In case, the quality system has production like data (i.e. was built from the productive system recently) and SDBI_CLUSTER_CHECK did not report errors it is normally fine to skip it for the productive system.

Can I run SDBI_CLUSTER_CHECK in parallel?

Yes, SDBI_CLUSTER_CHECK can be parallelized per table cluster only. R3check should be faster and could be manually parallelized, but you need to create the according R3check control files (Note 89384).

Where should I run the SDBI_CLUSTER_CHECK?

The job should be executed on the database server.

Does R3Check update the logs in SDBI_CLUSTER_CHECK?

No, it does not update the status of the SDBI_CLUSTER_CHECK.

Can I resume the report SDBI_CLUSTER_CHECK?

Yes, if you are using the last version of the report, you can do it by Select option "Execute Cluster Check → tab "Restart Cancelled Cluster check Run" and select the ID as show in the screenshot:

 

Results from report SDBI_CLUSTER_CHECK show initial cluster records for table X, what to do?

In this case, there are inconsistent entries in your cluster, you must contact the relevant table owner to find how to fix it. For example, if the cluster is EDI30C, the component is BC-MID-ALE. If you are not sure how to find the relevant component, we recommend to read the following SAP Note:

SAP Note 1348055 contains the list of component of all clusters.



  • No labels