currently the container files of our ADABAS database are spread over 3 pubsets: :ASSO:, :DATA:, :WORK:
/FI :ASSO:$ADA01.ADA01.ASSO,LINK=DDASSOR1,OPEN=INOUT,BLKSIZE=(STD,3),- / SHARUPD=YES /FI :DATA:$ADA01.ADA01.DATA,LINK=DDDATAR1,OPEN=INOUT,BLKSIZE=(STD,5),- / SHARUPD=YES /FI :WORK:$ADA01.ADA01.WORK,LINK=DDWORKR1,OPEN=INOUT,BLKSIZE=(STD,5),- / SHARUPD=YES /FI :DSIM:$ADA01.ADA01.DSIM,LINK=DDDSIMR1,BLKS=(STD,15),OPEN=INOUT,- / SHARUPD=YES
We’re planning to consolidate all pubsets on new devices to one single pubset.
This of course will affect our database nucleus jobs. Our action plan is as follows:
- stop nucleus
- ADASAV offline entire database,
- $ARCHIVE EXPORT of all pubsets including database container files,
- $ARCHIVE RESTORE into new single pubset,
- adapt nucleus jobs to reflect the new location of the container files
- restart nucleus again
- nucleus is active (hopefully)
The device type (2004) will not change.
Is this approach valid, what do the experts say.
Any input is appreciated …