|
View previous topic :: View next topic
|
| Author |
Message |
arunn
New User
Joined: 03 Sep 2010
Posts: 15
Location: Bengaluru
|
|
|
|
Re-posting as the previous post was posted in a non-syncsort forum. Apologies if it seems a duplicate.
I have a unique problem with how syncsort is working. My shop uses 2 mainframe environments - one runs on OS390 while other runs on Z/os 2 .
The job which performs sort and so far seems to be working perfectly in one mainframe system however the same job with same code is giving an error in different environment. The only visible difference i see is with the Syncsort for Z/os Version.
Below is the code snippet. Trying to understand what might possibly be wrong/incompatible here.
Log from the Code that works ----
| Code: |
SYNCSORT FOR Z/OS 2.1.4.1R
SYSIN :
SORT FIELDS=(13,31,A,211,27,A,5,7,A),FORMAT=BI
WER108I SORTIN : RECFM=VB ; LRECL= 654; BLKSIZE= 27998
WER073I SORTIN : DSNAME=INPUT.DATASET
WER110I SORTOUT : RECFM=VB ; LRECL= 654; BLKSIZE= 27998
WER074I SORTOUT : DSNAME=OUTPUT.DATASET
WER036B G=1155
WER147I CONTROL FIELD GREATER THAN RECORD LENGTH, POSSIBLE OUT OF SEQUENCE RECORD
WER177I TURNAROUND SORT PERFORMED |
Log from the same code in a different Mainframe environment that's failing.
| Code: |
SYNCSORT FOR Z/OS 2.1.0.0R
SYSIN :
SORT FIELDS=(13,31,A,211,27,A,5,7,A),FORMAT=BI
WER108I SORTIN : RECFM=VB ; LRECL= 654; BLKSIZE= 27998
WER073I SORTIN : DSNAME=INPUT.DATASET
WER110I SORTOUT : RECFM=VB ; LRECL= 654; BLKSIZE= 27998
WER074I SORTOUT : DSNAME=OUTPUT.DATASET
WER462I OUTPUT LRECL DIFFERS FROM SORTOUT LRECL
WER036B G=633
WER027A CONTROL FIELD BEYOND RECORD |
Be Safe,
Thanks. |
|
| Back to top |
|
 |
sergeyken
Senior Member
Joined: 29 Apr 2008
Posts: 2286
Location: USA
|
|
|
|
Your input records do have format VB, and you have got a particular one with LRECL<238 - hence your sort fields fell outside of the record field.
First of all, you need to decide: what to do with those short records? When decided, various solutions can be found to implement. |
|
| Back to top |
|
|
arunn
New User
Joined: 03 Sep 2010
Posts: 15
Location: Bengaluru
|
|
|
|
There was an mistake in copying the output of the errored part - Posting the correct one here as not able to edit the one earlier and also the SORT step.
| Code: |
//*
//SORT EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD DSN=INPUT.DATASET,DISP=SHR <<<< LRECL=654,RECFM=VB
//SORTOUT DD DSN=OUTPUT.DATASET,
// DISP=(NEW,CATLG,DELETE),
// SPACE=(CYL,(10,10),RLSE),
// LRECL=654,RECFM=VB
//SYSIN DD *
SORT FIELDS=(13,31,A,211,27,A,5,7,A),FORMAT=BI
//* |
Log from the successful run.
| Code: |
SYNCSORT FOR Z/OS 2.1.4.1R
SYSIN :
SORT FIELDS=(13,31,A,211,27,A,5,7,A),FORMAT=BI
WER108I SORTIN : RECFM=VB ; LRECL= 654; BLKSIZE= 27998
WER073I SORTIN : DSNAME=INPUT.DATASET
WER110I SORTOUT : RECFM=VB ; LRECL= 654; BLKSIZE= 27998
WER074I SORTOUT : DSNAME=OUTPUT.DATASET
WER036B G=1155
WER147I CONTROL FIELD GREATER THAN RECORD LENGTH, POSSIBLE OUT OF SEQUENCE RECORD
WER177I TURNAROUND SORT PERFORMED |
Log from the same code in a different Mainframe environment that's failing.
| Code: |
SYNCSORT FOR Z/OS 2.1.0.0R
SYSIN :
SORT FIELDS=(13,31,A,211,27,A,5,7,A),FORMAT=BI
WER108I SORTIN : RECFM=VB ; LRECL= 654; BLKSIZE= 27998
WER073I SORTIN : DSNAME=INPUT.DATASET
WER110I SORTOUT : RECFM=VB ; LRECL= 654; BLKSIZE= 27998
WER074I SORTOUT : DSNAME=OUTPUT.DATASET
WER036B G=633
WER027A CONTROL FIELD BEYOND RECORD |
The record length of the input and output file being created is 654. No changes being done there on the length. I am confused about why its working in one environment and not the other? |
|
| Back to top |
|
|
sergeyken
Senior Member
Joined: 29 Apr 2008
Posts: 2286
Location: USA
|
|
|
|
I guess, you seriously need to RTFM.
With RECFM=VB, LRECL=654 each particular record can have any LRECL, between 4 and 654 bytes. Whenever your particular specific record is shorter than 238, you get the error you mentioned.
Again: RTFM, RTFM, and RTFM!!!
P.S.
Maybe, your different environments do have different SORT default options established, to ignore specific situations with records.
Anyway, you need to understand well all those details, starting from understanding LRECL when RECFM=VB... |
|
| Back to top |
|
|
sergeyken
Senior Member
Joined: 29 Apr 2008
Posts: 2286
Location: USA
|
|
|
|
If you read carefully the explanation of WER147W message (can you?) then you would easily find that SORT parameter VLTEST causes different behavior of the utility in those your two environments.
Why you are not able to read the manual before starting your screaming at the forum? Do you speak English? |
|
| Back to top |
|
|
Rohit Umarjikar
Global Moderator
Joined: 21 Sep 2010
Posts: 3109
Location: NYC,USA
|
|
|
|
Try
| Code: |
| OPTION COPY, VLSHRT |
Besides, Syncsort releases don’t look the same and in one you already have a warning and in second that warning turned into an error. The error one looks to be the older version of the synsort. |
|
| Back to top |
|
|
sergeyken
Senior Member
Joined: 29 Apr 2008
Posts: 2286
Location: USA
|
|
|
|
| Rohit Umarjikar wrote: |
Try
| Code: |
| OPTION COPY, VLSHRT |
|
VLSHRT is an option for DFSORT utility, not for SYNCSORT. |
|
| Back to top |
|
|
Joerg.Findeisen
Senior Member
Joined: 15 Aug 2015
Posts: 1442
Location: Bamberg, Germany
|
|
|
|
From the manual:
| Code: |
VLTEST=(n/ 1 , ON /OFF /OFF4) -> Indicates the type of validity testing to be done when processing variable-length records.
VLTESTI=n/ 0 -> Indicates action to be taken when a variable-length record does not contain all fields referenced by INCLUDE or OMIT processing. |
|
|
| Back to top |
|
|
sergeyken
Senior Member
Joined: 29 Apr 2008
Posts: 2286
Location: USA
|
|
|
|
| Joerg.Findeisen wrote: |
From the manual:
| Code: |
VLTEST=(n/ 1 , ON /OFF /OFF4) -> Indicates the type of validity testing to be done when processing variable-length records.
VLTESTI=n/ 0 -> Indicates action to be taken when a variable-length record does not contain all fields referenced by INCLUDE or OMIT processing. |
|
If TS is not able to RTFM, then it is useless to quote the manual at the forum...  |
|
| Back to top |
|
|
Rohit Umarjikar
Global Moderator
Joined: 21 Sep 2010
Posts: 3109
Location: NYC,USA
|
|
|
|
| sergeyken wrote: |
| Rohit Umarjikar wrote: |
Try
| Code: |
| OPTION COPY, VLSHRT |
|
VLSHRT is an option for DFSORT utility, not for SYNCSORT. |
That's correct.
If I remember VLTEST=1 is the default and equivalent of VLSHRT in DFSORT. So in SYNCSORT , if you provide VLSHRT then internally it will convert to VLTEST equivalent instead of a syntax error.
If TS make a use of PARM VLSTEST=1 then error should go away and on the side can as well try for OPTION COPY,VLSHRT. |
|
| Back to top |
|
|
sergeyken
Senior Member
Joined: 29 Apr 2008
Posts: 2286
Location: USA
|
|
|
|
| Rohit Umarjikar wrote: |
| sergeyken wrote: |
| Rohit Umarjikar wrote: |
Try
| Code: |
| OPTION COPY, VLSHRT |
|
VLSHRT is an option for DFSORT utility, not for SYNCSORT. |
That's correct.
If I remember VLTEST=1 is the default and equivalent of VLSHRT in DFSORT. So in SYNCSORT , if you provide VLSHRT then internally it will convert to VLTEST equivalent instead of a syntax error.
If TS make a use of PARM VLSTEST=1 then error should go away and on the side can as well try for OPTION COPY,VLSHRT. |
The printed explanation of WER147I message clearly points to VLTEST parameter. The only thing the TS needs to do is, read the manual about this message found in his own log. Plus:<deleted by mod> |
|
| Back to top |
|
|
|
|
  |
All times are GMT + 6 Hours |
|
