Portal | Manuals | References | Downloads | Info | Programs | JCLs | Master the Mainframes
IBM Mainframe Computers Forums Index
 
Register
 
IBM Mainframe Computers Forums Index Mainframe: Search IBM Mainframe Forum: FAQ Memberlist Usergroups Profile Log in to check your private messages Log in
 

 

Problem with writing to VSAM dataset
Goto page 1, 2  Next
 
Post new topic   Reply to topic    IBMMAINFRAMES.com Support Forums -> JCL & VSAM
View previous topic :: :: View next topic  
Author Message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Wed May 02, 2007 7:44 pm    Post subject: Problem with writing to VSAM dataset
Reply with quote

In this code from c++ programming guide the author uses DUMMY.DATA
and no word about what should be inside. Any ideas?
//REPRO EXEC PGM=IDCAMS
//SYSPRINT DD SYSOUT=*
//SYSIN DD *
REPRO INDATASET(userid.DUMMY.DATA) -
OUTDATASET(userid.KSDS.CLUSTER)
/*

Is this code the same?
//SYSPRINT DD SYSOUT=*
//SYSIN DD DUMMY
REPRO INDATASET(userid.my.test) -
OUTDATASET(userid.KSDS.CLUSTER)
But after this fopen works fine but creating of BLDIndex failed. Any ideas?
Back to top
View user's profile Send private message

expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8593
Location: Back in jolly old England

PostPosted: Wed May 02, 2007 7:51 pm    Post subject:
Reply with quote

You can not build an alternate index on an empty file.

I usually build two dummy records of '00'X & 'FF'X for the whole length of the record and repro them into the base cluster. Then perform the BUILDIX afterwards.
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Wed May 02, 2007 7:57 pm    Post subject:
Reply with quote

thanks. Can you tell what "two dummy records of '00'X & 'FF'X" Don't really know how this member should look like?
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Wed May 02, 2007 8:01 pm    Post subject:
Reply with quote

Sorry, i mean that i don't know what should be inside this dummy records. I am new to this subject. Can you expaing in more detail?
Back to top
View user's profile Send private message
expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8593
Location: Back in jolly old England

PostPosted: Wed May 02, 2007 8:20 pm    Post subject:
Reply with quote

You know how long the records are for the base cluster, because you need that info to define the cluster.

In an external dataset, set up two records equal to the maximum record length of the file, and for each byte us '00'X HEX zeros - low values, and one record with 'FF'X - HEX FF - high values.

These populate the VSAM cluster, which in effect initialises it making it totally usable.

I'll try and post some REXX code tomorrow, I've posted something very similar on another forum, so will try and dig it up either from there or from my own library, if I kept a copy. It will initialise any VSAM KSDS just by knowing the cluster name.
Back to top
View user's profile Send private message
expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8593
Location: Back in jolly old England

PostPosted: Thu May 03, 2007 4:51 pm    Post subject:
Reply with quote

Here's the JCL code to run the REXX. It executes an ISPF environment in batch. Because I trip from site to site where standards differ I use loads of symbolics so that the code executes immediately at any site after initial set up.

The REXX will populate the cluster with two records, one with a key value of HEX low values '00'X and one with HEX high values 'FF'X. Obviously the code can be changed if there are any specific requirements for header/trailer information to be inserted.

You can also change the length of the WORK DD dataset to accomodate records larger than the 2004 specified.
Code:

//*                                                                 
// SET     PREFIX='FX.',                                             
//         SYSTID=''                                                 
//*                                                                 
//         JCLLIB ORDER=(&PREFIX.&SYSUID..&SYSTID.TEMPISPF)         
//*                                                                 
//BATCH    EXEC ISPFBAT,PARM='%VSAMPOP &PREFIX.. &SYSUID.. &SYSTID..'
//WORK     DD DSN=&PREFIX.&SYSUID..&SYSTID.WORK,                     
//            DISP=(,CATLG,CATLG),                                   
//            SPACE=(TRK,1),                                         
//            RECFM=V,LRECL=2004                                     
//VSAMFLS  DD *                                                     
Vsam cluster name 001
Vsam cluster name 002
Vsam cluster name 003
/*                                                                   
//DELETE   EXEC PGM=IEFBR14,COND=EVEN                               
//DD01     DD DSN=&PREFIX.&SYSUID..&SYSTID.WORK,                     
//            DISP=(MOD,DELETE),SPACE=(TRK,0)                       


And the REXX code executed is
Code:

/* REXX ** POPULATE SELECTED VSAM KSDS WITH HIGH AND LOW HEX KEYS    */

Z = MSG('OFF')
ARG PF US SI .
IF PF = '.' THEN PF = ''
IF US = '.' THEN US = ''
IF SI = '.' THEN SI = ''

WORKDSN = STRIP(PF!!US!!SI!!"WORK")

S1 = "KEYLEN"
S2 = "RKP"
S3 = "REC-TOTAL"

"EXECIO * DISKR VSAMFLS ( STEM FLS. FINIS"
DO A = 1 TO FLS.0
  PARSE VAR FLS.A DSN
  DSN = STRIP(DSN)

  CX1 = 0
  CX2 = 0
  CX3 = 0

  X=OUTTRAP(LST.)
  " LISTC ENT('"DSN"') ALL"
  X=OUTTRAP(OFF)

  DO B = 30 TO LST.0

    IF CX1 <> 1 THEN IX1 = POS(S1,LST.B)
    IF CX2 <> 1 THEN IX2 = POS(S2,LST.B)
    IF CX3 <> 1 THEN IX3 = POS(S3,LST.B)

    IF IX1 <> 0 & CX1 <> 1 THEN DO
      KLN = STRIP(TRANSLATE(SUBSTR(LST.B,IX1+6,18),' ','-'))
      CX1 = 1
    END
    IF IX2 <> 0 & CX2 <> 1 THEN DO
      KPS = STRIP(TRANSLATE(SUBSTR(LST.B,IX2+6,18),' ','-'))
      CX2 = 1
    END
    IF IX3 <> 0 & CX3 <> 1 THEN DO
      RCS = STRIP(TRANSLATE(SUBSTR(LST.B,IX3+9,15),' ','-'))
      CX3 = 1
    END

    IF CX1 & CX2 & CX3 THEN DO
      B = LST.0
      IF RCS <> 0 THEN DO
        SAY "RECORDS COUNT GT 0 FOR "DSN
        ITERATE A
      END
      "FREE  FI(WORK,VSAM)"
      "ALLOC FI(WORK) DA('"WORKDSN"') SHR"
      "ALLOC FI(VSAM) DA('"DSN"') SHR"
      IF KPS > 0 THEN AA = LEFT(' ',KPS)!!COPIES('00'X,KLN)
       ELSE AA = COPIES('00'X,KLN)
      PUSH AA
      "EXECIO 1 DISKW WORK"
      IF KPS > 0 THEN AA = LEFT(' ',KPS)!!COPIES('FF'X,KLN)
       ELSE AA = COPIES('FF'X,KLN)
      PUSH AA
      "EXECIO 1 DISKW WORK ( FINIS"
      "FREE  FI(WORK)"
      "ALLOC FI(WORK) DA('"WORKDSN"') SHR"
      " REPRO IFILE(WORK) OFILE(VSAM) "
      SAY "RETURN CODE "RIGHT(RC,4)" FOR "DSN
      "FREE  FI(WORK,VSAM)"
      ITERATE A
    END
  END
END
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Thu May 03, 2007 5:01 pm    Post subject:
Reply with quote

Thank you. Have'n you got a code in c++? icon_question.gif
Back to top
View user's profile Send private message
expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8593
Location: Back in jolly old England

PostPosted: Thu May 03, 2007 5:04 pm    Post subject:
Reply with quote

Quote:
Thank you. Have'n you got a code in c++?

Reply best left blank !!!
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Thu May 03, 2007 5:37 pm    Post subject:
Reply with quote

Sorry if I asked something wrong. I am new to this. may be you recommend some books about creating vsam. I have only DFSMS using data sets and c++ user's guide. And may be there is a way for more easier way to create this dummy data set? icon_neutral.gif
Back to top
View user's profile Send private message
expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8593
Location: Back in jolly old England

PostPosted: Thu May 03, 2007 5:46 pm    Post subject:
Reply with quote

Quote:
Sorry if I asked something wrong.

Always difficult when it's not your first language icon_confused.gif

For creating VSAM use this link

http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/DGT2I230/CCONTENTS?DT=20040628101835


For even more information on VSAM, try this link

http://www.redbooks.ibm.com/redbooks/pdfs/sg246105.pdf


As for C++ ....................... never used it
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Thu May 03, 2007 5:50 pm    Post subject:
Reply with quote

To copy
“dummy” indexed-sequential records (with X'FF' in the first byte), specify the
DUMMY option in the ENVIRONMENT parameter.

Found in using data sets? may be i should use this? icon_biggrin.gif
Back to top
View user's profile Send private message
expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8593
Location: Back in jolly old England

PostPosted: Thu May 03, 2007 6:00 pm    Post subject:
Reply with quote

Give it a try ....................... but I have always loaded my own records into a KSDS to initialise it. See if it works. It will work, I'm sure otherwise IBM would not tell you to do it.

The REXX above is a quick and easy way to initialise the cluster without having to know anything about the cluster except its name.

To initialise it properly, the key should be populated - that's why I use '00'X & 'FF'X because these are the lowest and highest possible values you could use.
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Thu May 03, 2007 9:32 pm    Post subject:
Reply with quote

Can anybody write a small jcl which creates dummy data set for initializing VSAM and used in repro?
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Mon May 07, 2007 2:03 pm    Post subject:
Reply with quote

any ideas? how to create this dummy data set using jcl? really need help icon_exclaim.gif
Back to top
View user's profile Send private message
muthuvel

Active User


Joined: 29 Nov 2005
Posts: 217
Location: Chennai

PostPosted: Mon May 07, 2007 2:40 pm    Post subject: Re: Problem with writing to VSAM dataset
Reply with quote

You can try VSAMINIT if it is available in your organisation.
This is to initialize the VSAM file with low values.
Code

Code:
//R020     EXEC PGM=VSAMINIT,
//INIT01   DD  DSN=<VSAM file>,     
//             DISP=SHR                             
//SYSOUT   DD  SYSOUT=*                             
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Mon May 07, 2007 3:33 pm    Post subject:
Reply with quote

IEFC621I EXPECTED CONTINUATION NOT RECEIVED

unfortunately not available. I can't understand what should be inside.
000000
000000 and where
fffff
fffff
?
And can't find in documentation. no word about it. in all examples they use something like userid.dummy
Back to top
View user's profile Send private message
expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8593
Location: Back in jolly old England

PostPosted: Mon May 07, 2007 5:06 pm    Post subject:
Reply with quote

Aleksey

Quote:
To copy
?dummy? indexed-sequential records (with X'FF' in the first byte), specify the
DUMMY option in the ENVIRONMENT parameter.

Indexed-sequential is a completely different access method from VSAM, and as far as I am aware is no longer supported by IBM, being replaced by VSAM.

Quote:
unfortunately not available. I can't understand what should be inside.
000000
000000 and where

Once a VSAM file has been defined, it can not be immediately used by any program until it has been initialised. This can be done by the first program to access the file, but that is not how I usually do this.

You know the offset of the key within the record, so add 1 to the offset to get the real position of the record.

Let us assume it is 80 bytles long with the key at offset 0 (position 1), for a length of 12.

Create a dataset that is 80 bytes long, and edit it by creating a blank line. Then on the command line type HEX ON
Code:

''''''                                                                 
       44444444444444444444444444444444444444444444444444444444444444444
       00000000000000000000000000000000000000000000000000000000000000000


Now, to create a key of '00'X overtype the all of the 4's from the top line with 0 (zero), and then type SAVE.

Code:

''''''                                                                 
       00000000000044444444444444444444444444444444444444444444444444444
       00000000000000000000000000000000000000000000000000000000000000000


Type HEX OFF, and you will then see
Code:

0000001


So now you have built the data to initialise your KSDS. I for one also generate a 'FF'X as well. So create a new line in your dataset and this time with HEX ON, overtype the 4's in the top line with F, and also each 0 below the 4.
Code:

000001                                   
       0000000000004444444444444444444444
       0000000000000000000000000000000000
-----------------------------------------
000002                                   
       FFFFFFFFFFFF4444444444444444444444
       FFFFFFFFFFFF0000000000000000000000

Type SAVE

Now, when you REPRO this dataset to the VSAM KSDS it is fully initialised with two records, one containing the lowest possible value in the key field, and one with the highest possible value.

Give it a go with your VSAM dataset, and if you still have problems post back here.
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Mon May 07, 2007 10:28 pm    Post subject:
Reply with quote

Some errors:

Code:
IDC1645I NONUNIQUE AIX KEY 4040404040404040   
         PRIME KEY IS 00000000               
IDC1645I NONUNIQUE AIX KEY 4040404040404040   
         PRIME KEY IS FFFFFFFF               
IDC1653I xxxxxxx.KSDS.UAIX BUILT WITH ERRORS


Code:
IDC1644I ALTERNATE INDEX KEY NOT IN BASE RECORD 00000000 
IDC1644I ALTERNATE INDEX KEY NOT IN BASE RECORD FFFFFFFF 
IDC1653I xxxxxxx.KSDS.NUAIX BUILT WITH ERRORS
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Mon May 07, 2007 10:55 pm    Post subject:
Reply with quote

Problem was solved. Thank you very much. icon_biggrin.gif
Back to top
View user's profile Send private message
Aleksey

New User


Joined: 24 Apr 2007
Posts: 13
Location: Russia

PostPosted: Tue May 08, 2007 12:57 am    Post subject:
Reply with quote

When I write to vsam i see this error
Code:
Reason code = 8                 
RBA = 0                         
Last op = 115       


Do you know what does it mean and where I can find the description of this error? in which book?
Back to top
View user's profile Send private message
View previous topic :: :: View next topic  
Post new topic   Reply to topic    IBMMAINFRAMES.com Support Forums -> JCL & VSAM All times are GMT + 6 Hours
Goto page 1, 2  Next
Page 1 of 2

 

Search our Forum:

Similar Topics
Topic Author Forum Replies Posted
No new posts FTP - JCL failed while passing FTP co... Suneetha1612 JCL & VSAM 12 Wed Nov 16, 2016 7:33 pm
No new posts Problem in writing Output file vickey_dw COBOL Programming 5 Mon Nov 14, 2016 11:14 pm
No new posts BWO option in VSAM blayek CICS 3 Sat Nov 05, 2016 10:47 am
No new posts Problem in Running Query via JCL vickey_dw DB2 3 Tue Oct 18, 2016 11:11 pm
No new posts To trim spaces in a pipe delimited da... Nish84 COBOL Programming 16 Mon Oct 10, 2016 1:54 pm


Facebook
Back to Top
 
Mainframe Wiki | Forum Rules | Bookmarks | Subscriptions | FAQ | Tutorials | Contact Us