View previous topic :: View next topic
Author
Message
vina2010 New User Joined: 06 Sep 2010Posts: 19 Location: Bangalore
Hi All
We have been trying to run a JOINKEYS SORT Step for two files. But jobs are cancelled by operator saying it is occupying almost 100% SPOOL and they cancel it. They are not sure what is causing the SPOOL to fill from our job.
But when I looked into this spool below PARMS are listed in the spool
JESMSGLG
JESJCL
JESYSMSG
SYSPRINT
SYSOUT
JNF1JMSG
JNF2JMSG
CAIPRINT
ABENDAID
SORTSNAP
Only SORTSNAP is what I can see is increasing. is that causing the problem? How do I fix this? When I read about SORTSNAP in google it says its a dynamically allocated DD step from SORT and should not be specified. We have not specified it. Please help.
FILE2 has around 2 million records is that causing a problem.
JCL below
Code:
//S01SORT EXEC PGM=SORT,COND=(8,LT)
//SORTJNF1 DD DSN=FILE1,DISP=SHR
//SORTJNF2 DD DSN=FILE2,DISP=SHR
//BOTH DD DSN=OUTPUT,
// DISP=(,CATLG,DELETE),SPACE=(TRK,(75,30),RLSE),
// DCB=(RECFM=FB,LRECL=620)
//SORTOUT DD SYSOUT=*
//SYSABEND DD SYSOUT=A
//SYSPRINT DD SYSOUT=A
//SYSUDUMP DD SYSOUT=A
//SYSOUT DD SYSOUT=A
//SYSIN DD *
JOINKEYS FILE=F1,FIELDS=(1,7,A)
JOINKEYS FILE=F2,FIELDS=(19,7,A)
JOIN UNPAIRED,F1,F2
REFORMAT FIELDS=(F2:1,620,F1:1,7,?)
OPTION COPY
OUTFIL FNAMES=BOTH,INCLUDE=(628,1,CH,EQ,C'B'),
BUILD=(1,620)
/*
Back to top
Bill Woodger Moderator Emeritus Joined: 09 Mar 2011Posts: 7309 Location: Inside the Matrix
75 tracks primary space is not a lot for 2.5m 620-byte records.
Post these in full please, in the Code tags.
SYSOUT
JNF1JMSG
JNF2JMSG
And the first 20 of these
CAIPRINT
ABENDAID
First 10 and last 10 of this.
SORTSNAP
Why are you processing UNPAIRED,F1,F2, but have only one OUTFIL which only contains matched records?
Why include the F2 data in the REFORMAT if it is not used?
Do you have duplicate keys? Sounds like you might have a lot.
Back to top
dick scherrer Moderator Emeritus Joined: 23 Nov 2006Posts: 19244 Location: Inside the Matrix
Hello and welcome to the forum,
Suggest you change SORTOUT to a dataset . . .
The SORTOUT DD SYSOUT=* could fill the spool . . .
It will help when you provide the requested info.
Back to top
Bill Woodger Moderator Emeritus Joined: 09 Mar 2011Posts: 7309 Location: Inside the Matrix
Good spot Dick.
With the SORTOUT DD and the OUTFIL, there are two copies of the data. One in the spool, and one fractured over however many packs.
Each record in the spool is going to take up five lines. That's a minimum of 12.5m. Some duplicates, and that goes way up.
Perhaps not noticed due to different class for the output?
Back to top
Please enable JavaScript!