A B37-04 says you did not successfully get all the space needed for the primary
allocation on the 1st volumes. Products like DTS Software, STOPX37, etc can
also help prevent these types of failures.
When SPACE= is coded you have 5 extents to get the primary allocation on the
volume. The secondary space allocation has upto (I think) 123 extents to get
it. But when you go after the primary allocation, you need to do it in 5
extents. This can be avoided by codind VOLUME COUNT in Data class, or using a
software product like STOPX37.
When the pack or pool is fragmented, small chunks of storage is available, then
it is possible you can not get 2000 cylinders in 5 extents. Hence the SB37-04
To correct this you can code your Data Class to allow more volumes (canidates)
to be available to your files.
In one of my Data Class definitions I have
Data Class Name : DB2VOLS
Description : MULTI VOLUMES FOR DB2 LINEAR DATA SETS
Recfm . . . . . . . . . :
Lrecl . . . . . . . . . :
Space Avgrec . . . . . . :
Avg Value . . . . :
Primary . . . . . :
Secondary . . . . :
Directory . . . . :
Retpd Or Expdt . . . . . :
Volume Count . . . . . . : 50
Add'l Volume Amount . . :
The VOLUME COUNT tells SMS to use upto 50 more volumes before failure.
Hope this helps.
Lizette
>Hi Kees,
>Thanks for this. But what is meant by "Your ACS routines appanrently don't
>provide you with a multivolume dataset"
>
>John
>
>
>> > We have IBMUSER.NDSP.** defined as SMS Managed datasets.
>> >
>> > My DBA trying to allocate SMS Managed dataset with space as
>> > //SYSREC01 DD DSN=IBMUSER.NDSP.DATA,
>> > // DISP=(NEW,CATLG,DELETE),
>> > // SPACE=(CYL,(2000,1800),RLSE)
>> >
>> > The job is failing with error
>> > IEC030I
>> >
>> B37-04,IFG0554A,IBMUSE1,S02,SYSREC01,9D96,ND0016,E6002130,IBMUSER.NDSP.D
>> ATA
>> >
>> > But if DBA submit the same job with space parameters as
>> >
>> > //SYSREC01 DD DSN=IBMUSER.NDSP.DATA,
>> > // DISP=(NEW,CATLG,DELETE),
>> > // SPACE=(CYL,(2000,1800),RLSE),
>> > // UNIT=(DISK,40)
>> >
>> > Then the JOB completes with RC 00.
>> >
>> > Please note that the only diff is of UNIT parameter. Can anyone
>> explain me
>> > why this happens ?
>> >
>> > John
>>
>> Yes, the difference is not the "DISK", but the "40". Your ACS routines
>> appanrently don't provide you with a multivolume dataset.
>>
>> Kees.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
