I have the request to convert from PDS loadlibsto PDSE loadlibs
But I have some doubts:
1) The BSTXCOPY is the utility recommended,isn't it?
2) Which one is the recommended code in theexec statment?
//COPYALL EXEC PGM=BSTXCOPY or
//COPYALL EXEC PGM=NDVRC1,PARM=BSTXCOPY
4) Can i have mixed libraries, PDS and PDSE,for a same ENVIRONMENT in difherent stages?
3) Members can be copied from PDS to PDSE and viceversa with move Endevoraction?
BSTCOPY is strickly a job run in a processor. BSTXCOPY was created to allow the job to be run outside of Endevor.
I would reccomend that is you can, run the job outside of Endevor. you cannot run BSTCOPY outside of Endevor and
BSTXCOPY cannot run in a processor. the jcl for BSTXCOPY can be found in prefix.qual.CSIQJCL,
if you have any problems you can give me a call (508-628-8971) or open a case in support.online.
Here are additional answers to your other questions:
2) Which one is the recommended code in the exec statment?
Use the second one.
If you invoke BSTXCOPY directly, it produces some Informational error messages C1G0700I and C1G0701I
Here is a link to a TEC doc that explains more --
4) Can i have mixed libraries, PDS and PDSE, for a same ENVIRONMENT in difherent stages?
Yes, you can.
3) Members can be copied from PDS to PDSE and viceversa with move Endevor action?
Yes, they can
Hi Team, CA SUPPORT TEAM has help me to figure out the previous topic, now I ask for your help if you know a recommendation for setting blksize for loadlibs with Lrecl = 0.
Hope you can help me
TEAM, thanks for your updates, here is a new one:
I'm coding BSTXCOPY as follows:
//COPY01 EXEC PGM=BSTXCOPY,COND=(0,LT,DELE01) //SYSPRINT DD SYSOUT=* //IND01 DD DSN=LNDV.NOPUT.UNITTEST.B.RLS.B.LOADLIB,DISP=SHR //OUT01 DD DSN=LNDV.UTIL.UNITTEST.B.RLS.B.LOADLIB.V2NEW, // LIKE=LNDV.NOPUT.UNITTEST.B.RLS.B.LOADLIB, // DISP=(NEW,CATLG,DELETE),DSNTYPE=LIBRARY //SYSIN DD * COPY INDD=((IND01,R)),OUTDD=OUT01
******June, I have coded also BSTXCOPY as NDVRC1 parm *******
but I'm getting the following message
C1X0511I COPY INDD=((IND01,R)),OUTDD=OUT01 C1X0511I IND01 = LNDV.NOPUT.UNITTEST.B.RLS.B.LOADLIB C1X0511I OUT01 = LNDV.UTIL.UNITTEST.B.RLS.B.LOADLIB.V2NEW C1X0507E THE BLKSIZE OF THE INPUT DATASET IS > THE BLKSIZE OF THE OUTPUT DATASE
These are the input and output data sets attributes
Enter "/" to select action Dsorg Recfm Lrecl Blksz----------------------------------------------------------------------LNDV.NOPUT.UNITTEST.B.RLS.B.LOADLIB PO U 0 23200LNDV.UTIL.UNITTEST.B.RLS.B.LOADLIB.V2NEW PO-E U 0 0
Does the LIKE parameter work out whit BSTXCOPY utility?Why the blksize output dataset is equal zero?Am I missing something?
From z/OS DFSMS Using Data Sets guide, IBM said The significance of the block size keyword (BLKSIZE) is slightly different from a PDS.You can get more detail in the Chapter 'Structure of a PDSE' from this guide.
I suggest you the following IBM doc;Block Size (BLKSIZE) : www-01.ibm.com/support/knowledgecenter/SSLTBW_1.13.0/com.ibm.zos.r13.idad400/blksz.htm%23blksz* where there is a link to 'Structure of a PDSE'PDSE Logical Block Size: www-01.ibm.com/support/knowledgecenter/SSLTBW_2.1.0/com.ibm.zos.v2r1.idad400/d4290.htmAlso, you can ask you DFSMS administrator about the ACS routine that are already or could be implemented. Anyway, he should be able to give you some guideline.
I would always go half track blocking at 27998 but there is enough performance information on the internet that 6144 is optimal for batch libraries and 13030 is best for CICS.
You have to decide whether optimum blocking on the disk is important over the way that storage is read in.
I can only suggest you do your own research and consult with your application teams.