Terry,
The comment on secondary allocation is incorrect, unless you are talking
about SMS Extended datasets with Space Constraint Relief. Those
datasets will allow secondary allocation to be allocated in multiple
extents - other datasets require the secondary allocation to be
satisfied by a single extent on the volume, and extension fails if no
available free extent exists.
The other thing anyone writing JCL needs to keep in mind is that there
is no guarantee that any secondary allocations will be possible.
Availability of space for secondary allocations is not a requirement
when volumes are assigned to a dataset. For ordinary sequential or PDS
datasets, there must be a volume on which primary allocation can be
satisfied in 5 or fewer extents or the job will fail before the step
initiates. Once all primary allocations succeed, the step can start;
but if space extension is required during execution you could get
anywhere from 0 to 15 secondaries before you either run out of space on
the volume or reach max extents on the dataset. It's basically a crap
shoot on secondary allocations in a busy shop with high DASD utilization.
J C Ewing
Terry Sambrooks wrote:
Hi Larry,
In respect of you query "I always thought when you gave a secondary
space allocation that it went to 16 extents(if needed), but I am reading
that it now goes to 123. Is this true?? We are a non sms shop.
One more question, does the system check for all secondary allocations or
will it work 1 by 1 as needed?"
The answer is that it depends.
The JCL Reference states "The system can allocate up to 123 extents for a
data set on a volume if it is a PDSE, an HFS data set, an extended format
data set, or a VSAM data set in an ICF catalog."
However traditional sequential and partitioned data sets are still limited
to 16 extents per volume.
Note that there is a significant difference between secondary request and an
extent. If I code SPACE=(CYL,(10,5)) I may get my secondary request in one 5
cylinder extent if it is available. However I could just as easily get my 5
cylinders as five 1 cylinder extents on a fragmented pack. Without
additional volumes indicated this could lead to an unexpected B37.
Kind regards - Terry
Terry Sambrooks
Director
KMS-IT Limited
228 Abbeydale Road South
Dore, Sheffield, S17 3LA, UK
Tel: +44 (0)114 262 0933
WEB: www.legac-e.co.uk
Company Reg: 3767263 at the above address
...
----------------------------------------------------------------------
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
