IBM Mainframe Forum Index
 
Log In
 
IBM Mainframe Forum Index Mainframe: Search IBM Mainframe Forum: FAQ Register
 

SE37 Abend in Sort


IBM Mainframe Forums -> DFSORT/ICETOOL
Post new topic   Reply to topic
View previous topic :: View next topic  
Author Message
nagarajan.dharani

New User


Joined: 27 Dec 2006
Posts: 36
Location: Chennai

PostPosted: Tue Feb 19, 2008 12:53 pm
Reply with quote

Hi,
I am getting SE37 space abend while using the following sort job.

Code:

//STEP005  EXEC PGM=SORT
//SYSIN    DD  *
   OPTION COPY
   OUTFIL FNAMES=SORTOF1,
     INCLUDE=(1,1,CH,EQ,C'D'),
     OUTREC=(1,1,2,4,BI,TO=ZD,LENGTH=10,122X)
   OUTFIL FNAMES=SORTOF2,
     INCLUDE=(1,1,CH,EQ,C'A'),
     OUTREC=(1,1,2,4,BI,TO=ZD,LENGTH=10,122X)
   OUTFIL FNAMES=SORTOF3,
     INCLUDE=(1,1,CH,EQ,C'C'),
     OUTREC=(1,1,2,4,BI,TO=ZD,LENGTH=10,122X)
/*
//SYSOUT   DD SYSOUT=*
//SORTIN   DD DSN=input FILE       
//SORTOF1 DD DISP=(,CATLG,DELETE),
//             UNIT=SYSDA,
//             SPACE=(CYL,(10000,10000),RLSE),
//             DSN=OUTPUTFILE1,
//             VOL=(,,,20)
//SORTOF2 DD DISP=(,CATLG,DELETE),
//             UNIT=SYSDA,
//             SPACE=(CYL,(10000,10000),RLSE),
//             DSN=OUTPUTFILE2,
//             VOL=(,,,20)
//SORTOF3 DD DISP=(,CATLG,DELETE),
//             UNIT=SYSDA,
//             SPACE=(CYL,(10000,10000),RLSE),
//             DSN=OUTPUTFILE3,
//             VOL=(,,,20)


I'm getting the following error and if i increase the space also the error is
same.
ICE231I 0 STORAGE USED FOR OUTFIL : BELOW 16M = 28672, ABOVE ICE210I 0 SORTOF1 : EXCP USED, LRECL = 133, BLKSIZE = 133, TYPE
ICE210I 0 SORTOF2 : EXCP USED, LRECL = 133, BLKSIZE = 133, TYPE
ICE210I 0 SORTOF3 : EXCP USED, LRECL = 133, BLKSIZE = 133, TYPE
ICE751I 2 EF-K10929 CB-K90000 F0-Q84357 DA-K26318
ICE185A 0 AN SE37 ABEND WAS ISSUED BY DFSORT, ANOTHER PROGRAM OR AN EXIT (PHAS

The input file has more than 5 million records.

Can anyone help me out in this.

Thanks,
Dharani
Back to top
View user's profile Send private message
vasanthkumarhb

Active User


Joined: 06 Sep 2007
Posts: 275
Location: Bang,iflex

PostPosted: Tue Feb 19, 2008 1:34 pm
Reply with quote

Hello,

This is a space abend related to PDS used in ur JCL, Compress ur libraries and the dataset ur using in JCL will clear ur abend.
to do this go to 3.4 option write Z at the left most space of the data set press enter.
Back to top
View user's profile Send private message
expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8797
Location: Welsh Wales

PostPosted: Tue Feb 19, 2008 2:13 pm
Reply with quote

vasanthkumarhb wrote:
Hello,

This is a space abend related to PDS used in ur JCL, Compress ur libraries and the dataset ur using in JCL will clear ur abend.
to do this go to 3.4 option write Z at the left most space of the data set press enter.

I can not see any PDS mentioned in the above JCL - Can you tell me which one it is ?
I checked in the manual, and there was no indication that E37 is specific to PDS.

To nagarajan.dharani
Which DD name is showing the abend ?
Which DASD are you using, I would guess at least model 27s by your allocation requests. A very large allocation can cause more problems that it solves with fragmented DASD, so take a look at the DASD too.
Back to top
View user's profile Send private message
enrico-sorichetti

Superior Member


Joined: 14 Mar 2007
Posts: 10872
Location: italy

PostPosted: Tue Feb 19, 2008 2:28 pm
Reply with quote

Not strictly sort related, rather a more general storage management issue,

why are You using unblocked records, it' s the worst way of using dasds space

for a record lenght of 133 here are the percentages

Code:
blocking factor      blocks per track   records per track   device utilization
unblocked (133)      72                   72                 20.1%   
half track (27930)    2                  420                 98.8%


meditate [storage] people meditate icon_biggrin.gif
Back to top
View user's profile Send private message
expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8797
Location: Welsh Wales

PostPosted: Tue Feb 19, 2008 2:44 pm
Reply with quote

Enrico,

which book did you get the stats from.

I keep finding it, and then forgetting it again icon_eek.gif
Back to top
View user's profile Send private message
enrico-sorichetti

Superior Member


Joined: 14 Mar 2007
Posts: 10872
Location: italy

PostPosted: Tue Feb 19, 2008 2:49 pm
Reply with quote

Me too icon_biggrin.gif

do forever
... I keep forgetting,
... search the net
...every time I get a different url
end

this time I got it from
http://programmingstuff.googlepages.com/blksize.htm
and from
http://www.dtssoftware.com/Ref%20Guide9.01.pdf
Back to top
View user's profile Send private message
vasanthkumarhb

Active User


Joined: 06 Sep 2007
Posts: 275
Location: Bang,iflex

PostPosted: Tue Feb 19, 2008 3:02 pm
Reply with quote

Hi Expat.


Quote:
SE37 --- A DATASET OPENED FOR OUTPUT USED ALL SPACE AVAILABLE ON
THE CURRENT VOLUME, AND NO MORE VOLUMES WERE AVAILABLE:

- NOT ENOUGH VOLUMES WERE SPECIFIED FOR THE DATASET THROUGH THE 'SER', VOLUME COUNT, OR 'REF' SUBPARAMETER OF THE VOLUME PARAMETER. WHEN ALL THE VOLUMES WERE FILLED, THE PROGRAM ATTEMPTED TO WRITE ANOTHER RECORD.

- FOR A PARTITIONED DATASET ON A DIRECT ACCESS VOLUME, ALL SPACE WAS FILLED WHEN THE PROGRAM ATTEMPTED TO WRITE ANOTHER RECORD.

(A PARTITIONED DATASET CAN RESIDE ON ONLY ONE VOLUME.)

- FOR A PARTITIONED DATASET ON A DIRECT ACCESS VOLUME, 16 EXTENTS HAD BEEN USED WHEN THE PROGRAM ATTEMPTED TO WRITE ANOTHER RECORD; HOWEVER, NO VOLUME WAS AVAILABLE FOR DEMOUNTING.


This is wat i got it from internet.......yes he is not using any dataset even i said if he using data set like libraries he can do this icon_wink.gif
Back to top
View user's profile Send private message
superk

Global Moderator


Joined: 26 Apr 2004
Posts: 4652
Location: Raleigh, NC, USA

PostPosted: Tue Feb 19, 2008 4:58 pm
Reply with quote

Shouldn't the job use SMS for the output datasets, eliminating the concern for this problem for anyone other than the local storage administrator?
Back to top
View user's profile Send private message
expat

Global Moderator


Joined: 14 Mar 2007
Posts: 8797
Location: Welsh Wales

PostPosted: Tue Feb 19, 2008 5:20 pm
Reply with quote

Unfortunately, as Enrico has pointed out, the RECFM parameter is the most wasteful. SMS would accept the user specified values for this.
Back to top
View user's profile Send private message
Frank Yaeger

DFSORT Developer


Joined: 15 Feb 2005
Posts: 7129
Location: San Jose, CA

PostPosted: Tue Feb 19, 2008 10:10 pm
Reply with quote

Quote:
why are You using unblocked records, it' s the worst way of using dasds space


Dharani,

I agree with Enrico that this is your main problem. DFSORT would normally automatically use a large BLKSIZE for your OUTFIL data sets. The fact that it didn't might indicate that your site has changed DFSORT's shipped default from SDB=INPUT to SDB=NO. SDB=NO prevents DFSORT from using system determined BLKSIZEs for output data sets. SDB=NO is a VERY BAD CHOICE!! (Not sure if anything else could have caused the small BLKSIZE, like ACS routines.)

I ran your job with 6 million input records and SORTOFx data sets with SPACE=(CYL,(150,150),RLSE) and let it use SDB and it worked fine. (I didn't try a lower size for the space.) The BLKSIZE selected was 27930.

You can override your site's choice of SDB=NO by adding the following to your job:

Code:

//DFSPARM DD *
   OPTION SDB=YES
/*


If you still get the small BLKSIZE, then something else is causing it and you might want to try specifying BLKSIZE=27930 on the SORTOFx DD statements.
Back to top
View user's profile Send private message
nagarajan.dharani

New User


Joined: 27 Dec 2006
Posts: 36
Location: Chennai

PostPosted: Thu Feb 21, 2008 3:50 pm
Reply with quote

Hi Frank/Enrico,

After changing 'F' format to 'FB' and giving blksize

with the OPTION SBS=YES my job ran fine with a space of
SPACE=(CYL,(200,200),RLSE)


Thanks a lot for all your help..

Can you tell me what is the difference between UNIT=SYSDA and
UNIT=DISK...??

Thanks,
Dharani
Back to top
View user's profile Send private message
Anuj Dhawan

Superior Member


Joined: 22 Apr 2006
Posts: 6250
Location: Mumbai, India

PostPosted: Thu Feb 21, 2008 8:25 pm
Reply with quote

Hi,

UNIT=type indicates the type of device. The following are the most common:

Disk a disk data set; this unit designation is used when your data set must be put on a particular disk. Used in conjuction with SER=volume-name.

SYSDA a system disk data set; used with temporary data sets.

TAPE tells the system a reel tape will be used. TAPE80 tells the system a cartridge tape will be used. Eventually all tape files will be on cartridge tape.

Hope this helps.
Back to top
View user's profile Send private message
dick scherrer

Moderator Emeritus


Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix

PostPosted: Thu Feb 21, 2008 8:41 pm
Reply with quote

Hello,

Quote:
Can you tell me what is the difference between UNIT=SYSDA and
UNIT=DISK...??
These are specific to your system.

I'd suggest you talk with the storage management or the system configuration people who support your environment to learn the complete list and when each should be specified there.
Back to top
View user's profile Send private message
View previous topic :: :: View next topic  
Post new topic   Reply to topic View Bookmarks
All times are GMT + 6 Hours
Forum Index -> DFSORT/ICETOOL

 


Similar Topics
Topic Forum Replies
No new posts How to split large record length file... DFSORT/ICETOOL 7
No new posts JCL sort card - get first day and las... JCL & VSAM 9
No new posts ISAM and abend S03B JCL & VSAM 9
No new posts Sort First/last record of a subset th... DFSORT/ICETOOL 7
No new posts how to calculate SUM value for VB fil... DFSORT/ICETOOL 1
Search our Forums:

Back to Top