z/OS v11 NFM -628 on a create on a partitioned index

Tim Wilkins

z/OS v11 NFM -628 on a create on a partitioned index

I am testing converting a partitioned table (table controlled) to a UTS and changing the indexes not partitioned to be partitioned mostly for parallel processing and being able to reorg at the partition level.   The table has 16 partitions.

The existing clustering unique index which is partitioned has these columns

CLIENT_NBR

ACCT_NBR

One existing non-partitioned index is defined as unique on column ACCT_NBR.  When I attempt to create it as a unique partitioned index I get the -628.   I read the manual explanation and it didn't quite make sense.  

After some thought I believe what I need to do is drop the clustering unique index (client_nbr and acct_nbr) and create it as the clustering index and the other index create it as unique on acct_nbr.   That would make sense which would make it such that no multiple clint_nbr's would have the same acct_nbr.  

Does this make sense?  

 

Walter Janißen

AW: z/OS v11 NFM -628 on a create on a partitioned index
(in response to Tim Wilkins)
Hi Tim

A unique DPSI must include all partitioning columns.

Kind regards
Walter Janißen [standard_IBM+Champ+7+Yr+Analytics]

ITERGO Informationstechnologie GmbH
Anwendungsentwicklung
Technische Anwendungsarchitektur
Victoriaplatz 2
D-40198 Düsseldorf
[login to unmask email]<mailto:[login to unmask email]>

ITERGO Informationstechnologie GmbH
Vorsitzender des Aufsichtsrats: Christian Diedrich
Geschäftsführung: Dr. Bettina Anders (Vorsitzende),
Lothar Engelke, Ina Kirchhof, Dr. Michael Regauer
Sitz: Düsseldorf, Handelsregister: Amtsgericht Düsseldorf HRB 37996

Von: Tim Wilkins [mailto:[login to unmask email]
Gesendet: Donnerstag, 26. Oktober 2017 18:38
An: [login to unmask email]
Betreff: [DB2-L] - z/OS v11 NFM -628 on a create on a partitioned index


I am testing converting a partitioned table (table controlled) to a UTS and changing the indexes not partitioned to be partitioned mostly for parallel processing and being able to reorg at the partition level. The table has 16 partitions.

The existing clustering unique index which is partitioned has these columns

CLIENT_NBR

ACCT_NBR

One existing non-partitioned index is defined as unique on column ACCT_NBR. When I attempt to create it as a unique partitioned index I get the -628. I read the manual explanation and it didn't quite make sense.

After some thought I believe what I need to do is drop the clustering unique index (client_nbr and acct_nbr) and create it as the clustering index and the other index create it as unique on acct_nbr. That would make sense which would make it such that no multiple clint_nbr's would have the same acct_nbr.

Does this make sense?



-----End Original Message-----
Attachments

  • image001.png (2.6k)