Page 1 of 1

INZUTILB - sortworks

PostPosted: Tue Nov 20, 2012 8:30 pm
by Jorgen Amnestal
Hello all

I'm running a batch job using pgm INZUTILB, (in DB2 V10/NFM)
trying to unload a table from an image copy dataset.
Problem is, it abends; U1017, which according to the manual is a failure in a sub routine (probably SORT).

I have 5 SORTWORK datasets, but they don't seem to do the trick,
//SORTWK01 DD SPACE=(CYL,(100,300),RLSE)
//SORTWK05 DD SPACE=(CYL,(100,300),RLSE)

and in the output you can (also) see 3 temporary sortworks:
IGD101I SMS ALLOCATED TO DDNAME (T001WK01)
DSN (SYS12325.T134203.RA000.P950JGNP.R0926634 )
STORCLAS (STANDARD) MGMTCLAS ( ) DATACLAS (STANDARD)
VOL SER NOS= TEM320

and the error message
ICE278I 1 2 WORK DATA SETS WERE INSUFFICIENT TO COMPLETE THIS SORT SO 1 ADDITIONAL WERE USED

Is there something new in V10, that requires new/different DD-cards for INZUTILB?

best rgds
/Jorgen

Re: INZUTILB - sortworks

PostPosted: Tue Nov 20, 2012 8:55 pm
by Robert Sample
An ICE278I message is merely informational and does not indicate any problem occurred in your sort. From the MAC manual:
ICE278I
x WORK DATA SETS WERE INSUFFICIENT TO COMPLETE THIS SORT |SO y ADDITIONAL WORK DATA SETS WERE USED
|
Explanation:
|

DFSORT initially intended to use only x work data sets for this |sort application. However, the work space requirements for the sort |were unexpectedly larger than anticipated, so DFSORT used y additional |work data sets.
|
System action:
|

None.
|
Programmer response:
|

Verify that DFSORT has the correct file size information to allow |it to accurately calculate the initial amount of work space needed. | See messages ICE098I and ICE253I for additional information.
Your abend should have generated some messages in the job log about the precise problem -- but the ICE278I message is not going to be the cause of the problem (at least not directly, but if you cannot provide SORT with an accurate file size estimate then space related abends are quite likely).

Re: INZUTILB - sortworks

PostPosted: Wed Nov 21, 2012 5:39 pm
by Jorgen Amnestal
Hi, Robert

and thanks for your response.
I guess what I'm really anxious to know is why the program doesn't acknowledge the five //SORTWK I put in the JCL,
but allocate three others instead. Too small, even!

best rgds
/Jorgen

Re: INZUTILB - sortworks

PostPosted: Wed Nov 21, 2012 8:59 pm
by dick scherrer
Hello,

DB2 "magically" (and very quietly) invokes DFSORT "under the covers".

Suggest you look in the DFSORT documentatiion for further qualification of the diagnostic information generated.