View previous topic :: View next topic
|
Author |
Message |
smaru
New User
Joined: 22 Oct 2008 Posts: 49 Location: India
|
|
|
|
Hi,
I am tring to sort a tape file with record count 30 million records. The sysin has Sort fields and Sum fields.
I have been providing workspace through dynamic allocation
OPTION DYNALLOC=(SYSTF,30),FILSZ=E30000000
But my Job failed with message,
SORT CAPACITY EXCEEDED - RECORD COUNT 279890649
I have tried increasing the dynalloc from 30 to 50 still it failed with same message.
Can someone suggest on this. |
|
Back to top |
|
|
smaru
New User
Joined: 22 Oct 2008 Posts: 49 Location: India
|
|
|
|
The LRECL of the output file is 60 |
|
Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006 Posts: 19244 Location: Inside the Matrix
|
|
|
|
Hello,
You might want to run a job to count the records on the input. You mention 30 million records, but the sort has seen 279 million+ when the capacity was exceeded.
If you post the complete jcl and control statements it may help us help you.
If all else fails, you can split the file into several smaller files, sort the smaller files and then merge them back into a large sorted file.
Is the input record the same length as the output? |
|
Back to top |
|
|
Robert Sample
Global Moderator
Joined: 06 Jun 2008 Posts: 8696 Location: Dubuque, Iowa, USA
|
|
|
|
You're estimating 30 million records but you've run 279 million through SORT when it runs out of space ... maybe you should increase your estimate first? |
|
Back to top |
|
|
Frank Yaeger
DFSORT Developer
Joined: 15 Feb 2005 Posts: 7129 Location: San Jose, CA
|
|
|
|
Smaru,
You've said your tape has 30 million records, which is obviously wrong since the ICE046A shows that DFSORT processed about 270 million records before it ran out of work space. How many records does your input file really have?
I can't tell if DFSORT is able to determine the input filesize automatically from your tape management system or not. If it can then it will ignore your FILESZ estimate whereas if it can't it will use your way too small FILSZ estimate.
Try increasing the FILSZ estimate to be more accurate. If that doesn't work, add the following to your job to tell DFSORT to show the diagnostic messages:
//SORTDIAG DD DUMMY
Then rerun the job and send me all of the JES and DFSORT messages offline (yaeger@us.ibm.com) and I'll take a look. |
|
Back to top |
|
|
|