DISP=(MOD,DELETE)

[g***@ibm-main.lst]

The JCL fragment:

//STEP EXEC PGM=IEFBR14
//*
//SYSUT1 DD DSN=user.FOO.BAR.SMPLOG,DISP=(MOD,KEEP),
// VOL=SER=TSO026,
// UNIT=SYSALLDA

where 'user.FOO.BAR.SMPLOG' does not pre-exist on the volume
produces allocation messages:

IEF236I ALLOC. FOR MODKEEP STEP
IEF237I 3ECA ALLOCATED TO SYSUT1
IEF142I MODKEEP STEP - STEP WAS EXECUTED - COND CODE 0000
IEF285I user.FOO.BAR.SMPLOG KEPT
IEF285I VOL SER NOS= TSO026.
IEF373I STEP/STEP /START 2007142.1554

.. but VTOC listing does not show the data set. As I read
the " 12.19.11 Summary of Disposition Processing" in the JCL
RM, I believe it should be kept. When I change to DISP=(NEW,KEEP),
it fails with:

IGD17045I SPACE NOT SPECIFIED FOR ALLOCATION OF DATA SET

Is this working as expected? Should I be able to understand
this?

Thanks,
gil

[Chris Mason]

Paul

You got my attention with the subject line, DISP=(MOD,DELETE), since I used
to encourage students to consider using this "trick" in order to immunise
their JCL against irritating errors when they had usually started task
procedures relating to networking products and they didn't care whether or
not a data set existed when the task was started. The types of data sets
involved were for traces, logs and dumps.

Typically the started task procedure had an IEFBR14 step where the data sets
were definitely erased if they happened to exist - from, typically of
course, any previous execution - using the DISP=(MOD,DELETE) "trick". The
IEFBR14 step was then followed by the "business" step.[1]

What you show in fact is DISP=(MOD,KEEP).

An explanation - not at all necessarily *the* explanation since it's very
much a guess - for the IEF285I message is that, with DISP=MOD, logically the
data set is considered to exist and thus it can logically be "kept".
However, with, DISP=NEW, the data set is considered *not* to exist - and you
get another error presumably earlier in the processing sequence if it does
exist - and so, before it can be "kept", it must actually be created. In
order to be created, the SPACE parameters are needed.

So *I* understand it. Of course whether or not I understand it correctly is
quite another thing!

Chris Mason

[1] Of course the "business" step program needed *not* to have the SYST
attribute specified in any SCHEDxx member - but that's another story - see
topic "SYST" dated Sat, 25 Feb 2006 07:59:37 -0800.

----- Original Message -----
From: "Paul Gilmartin" <***@ibm-main.lst>
Newsgroups: bit.listserv.ibm-main
To: <IBM-***@BAMA.UA.EDU>
Sent: Wednesday, May 23, 2007 12:05 AM
Subject: DISP=(MOD,DELETE)

Post by g***@ibm-main.lst
//STEP EXEC PGM=IEFBR14
//*
//SYSUT1 DD DSN=user.FOO.BAR.SMPLOG,DISP=(MOD,KEEP),
// VOL=SER=TSO026,
// UNIT=SYSALLDA
where 'user.FOO.BAR.SMPLOG' does not pre-exist on the volume
IEF236I ALLOC. FOR MODKEEP STEP
IEF237I 3ECA ALLOCATED TO SYSUT1
IEF142I MODKEEP STEP - STEP WAS EXECUTED - COND CODE 0000
IEF285I user.FOO.BAR.SMPLOG KEPT
IEF285I VOL SER NOS= TSO026.
IEF373I STEP/STEP /START 2007142.1554
... but VTOC listing does not show the data set. As I read
the " 12.19.11 Summary of Disposition Processing" in the JCL
RM, I believe it should be kept. When I change to DISP=(NEW,KEEP),
IGD17045I SPACE NOT SPECIFIED FOR ALLOCATION OF DATA SET
Is this working as expected? Should I be able to understand
this?
Thanks,
gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Doug Fuerst]

Actually, this was recently explained to me by IBM. (MOD,DELETE)
indicates that the dataset might exist, if it does, then MOD it. If
it does not, then create it and then delete it. But SPACE would be
required in any case where the dataset did not exist. This would also
be true for (MOD,KEEP). In this case, if the dataset did not exist it
would be created, and KEPT, but not cataloged. SPACE would be
required if the dataset did not exist.

Doug

Post by Chris Mason
Paul
You got my attention with the subject line, DISP=(MOD,DELETE), since
I used to encourage students to consider using this "trick" in order
to immunise their JCL against irritating errors when they had
usually started task procedures relating to networking products and
they didn't care whether or not a data set existed when the task was
started. The types of data sets involved were for traces, logs and dumps.
Typically the started task procedure had an IEFBR14 step where the
data sets were definitely erased if they happened to exist - from,
typically of course, any previous execution - using the
DISP=(MOD,DELETE) "trick". The IEFBR14 step was then followed by the
"business" step.[1]
What you show in fact is DISP=(MOD,KEEP).
An explanation - not at all necessarily *the* explanation since it's
very much a guess - for the IEF285I message is that, with DISP=MOD,
logically the data set is considered to exist and thus it can
logically be "kept". However, with, DISP=NEW, the data set is
considered *not* to exist - and you get another error presumably
earlier in the processing sequence if it does exist - and so, before
it can be "kept", it must actually be created. In order to be
created, the SPACE parameters are needed.
So *I* understand it. Of course whether or not I understand it
correctly is quite another thing!
Chris Mason
snip>>>>>>>>>>>>>>>
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Doug Fuerst
Consultant
BK Associates
Brooklyn, NY
[phone] (Office)
[phone] (Fax)
[phone] (Cell)
***@bkassociates.net

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[R.S.]

Post by Doug Fuerst
Actually, this was recently explained to me by IBM. (MOD,DELETE)
indicates that the dataset might exist, if it does, then MOD it. If it
does not, then create it and then delete it. But SPACE would be required
in any case where the dataset did not exist. This would also be true for
(MOD,KEEP). In this case, if the dataset did not exist it would be
created, and KEPT, but not cataloged. SPACE would be required if the
dataset did not exist.

Actually, the explanation is clear, but false. <g>
Despite of what IBM said, just repeat Gil's test.
In fact, I did it already, and have to self-correct my previous mail:
The devil is in catalog.
If you specify unit and volser, SPACE parameter in not required.
If you don't specify unit 7 volser - that mean you want to use catalog, then SPACE is mandatory for non-existent dataset. Obviously SMS-managed dataset are cataloged, so my previous answer was "almost true" as IBM answer above <g>


BTW: Anytime I think that nothing in JCL can surprise me, I see new posting from Gil.
Gil, you are JCL entomologist, thank you for that!

Regards
--
Radoslaw Skorupka
Lodz, Poland


--
BRE Bank SA
ul. Senatorska 18
00-950 Warszawa
www.brebank.pl

Sd Rejonowy dla m. st. Warszawy
XII Wydzia Gospodarczy Krajowego Rejestru Sdowego,
nr rejestru przedsibiorców KRS 0000025237
NIP: 526-021-50-88
Wedug stanu na dzie 01.01.2007 r. kapita zakadowy BRE Banku SA (w caoci opacony) wynosi 118.064.140 z. W zwizku z realizacj warunkowego podwyszenia kapitau zakadowego, na podstawie uchwa XVI WZ z dnia 21.05.2003 r., kapita zakadowy BRE Banku SA moe ulec podwyszeniu do kwoty 118.760.528 z. Akcje w podwyszonym kapitale zakadowym bd w caoci opacone.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Chris Mason]

Radoslaw and Zaromil

Now that you, Zaromil, has actually gone and bothered to read the manual for
us all, what you. Radoslaw, said still needs a minor adjustment, namely:

"If you specify unit and volser, the *data set* is not required" - as long
as you don't go and do something silly like try to use it!

It seems that my original informant may also have misled me all this time
and that OLD or SHR would have been equally as acceptable as MOD.

Chris Mason

----- Original Message -----
From: "R.S." <***@BREMULTIBANK.COM.PL>
Newsgroups: bit.listserv.ibm-main
To: <IBM-***@BAMA.UA.EDU>
Sent: Wednesday, May 23, 2007 8:43 AM
Subject: Re: DISP=(MOD,DELETE)

...
If you specify unit and volser, SPACE parameter is not required.
If you don't specify unit & volser - that mean you want to use catalog, then
SPACE is mandatory for non-existent dataset. Obviously SMS-managed dataset
are cataloged, so my previous answer was "almost true" as IBM answer above
<g>

...
--
Radoslaw Skorupka
Lodz, Poland

----- Original Message -----
From: "TISLER Zaromil" <***@IT-AUSTRIA.COM>
Newsgroups: bit.listserv.ibm-main
To: <IBM-***@BAMA.UA.EDU>
Sent: Wednesday, May 23, 2007 8:54 AM
Subject: Re: DISP=(MOD,DELETE)

...

Post by g***@ibm-main.lst
----------------------------------------
If the system obtains unit and volume information for an OLD, MOD, or SHR
status, the data set is treated as if it exists, whether or not it is
physically on the device. ----------------------------------------
and the "4.6.2.1.8 Disposition Processing of Data Sets that Do Not Exist"
in
----------------------------------------
When VOLUME and UNIT Are Coded: When you code VOLUME and UNIT on the DD
statement, a JCL error will occur if the problem program attempts to open
the data set. Otherwise, the data set disposition depends on the DISP
normal
When the normal termination disposition is KEEP, the job log will show
that
the data set was kept. ----------------------------------------
Zaromil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[R.S.]

Post by Chris Mason
Radoslaw and Zaromil
Now that you, Zaromil, has actually gone and bothered to read the manual
for us all, what you. Radoslaw, said still needs a minor adjustment,
"If you specify unit and volser, the *data set* is not required" - as
long as you don't go and do something silly like try to use it!
It seems that my original informant may also have misled me all this
time and that OLD or SHR would have been equally as acceptable as MOD.

It depends. Try DISP=(MOD,CATLG). It *works* for non-existent dataset, when you also specify UNIT&VOLSER.
If you omit DSNAME, it ...will work, but in quite different manner: DD is treated as it would describe temporary dataset. In this case CATLG is *ignored* with warning message.
So, DSNAME is important, although it can be omitted. However in the last case I described above (temp dsn) it can be omitted independently of UNIT&VOLSER existence.
--
Radoslaw Skorupka
Lodz, Poland


--
BRE Bank SA
ul. Senatorska 18
00-950 Warszawa
www.brebank.pl

Sd Rejonowy dla m. st. Warszawy
XII Wydzia Gospodarczy Krajowego Rejestru Sdowego,
nr rejestru przedsibiorców KRS 0000025237
NIP: 526-021-50-88
Wedug stanu na dzie 01.01.2007 r. kapita zakadowy BRE Banku SA (w caoci opacony) wynosi 118.064.140 z. W zwizku z realizacj warunkowego podwyszenia kapitau zakadowego, na podstawie uchwa XVI WZ z dnia 21.05.2003 r., kapita zakadowy BRE Banku SA moe ulec podwyszeniu do kwoty 118.760.528 z. Akcje w podwyszonym kapitale zakadowym bd w caoci opacone.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Doug Fuerst]

I have a headache....

Post by R.S.

Post by Chris Mason
Radoslaw and Zaromil
Now that you, Zaromil, has actually gone and
bothered to read the manual for us all, what
"If you specify unit and volser, the *data set*
is not required" - as long as you don't go and
do something silly like try to use it!
It seems that my original informant may also
have misled me all this time and that OLD or
SHR would have been equally as acceptable as MOD.

It depends. Try DISP=(MOD,CATLG). It *works* for
non-existent dataset, when you also specify
UNIT&VOLSER. If you omit DSNAME, it ...will
work, but in quite different manner: DD is
treated as it would describe temporary dataset.
In this case CATLG is *ignored* with warning
message. So, DSNAME is important, although it
can be omitted. However in the last case I
described above (temp dsn) it can be omitted
independently of UNIT&VOLSER existence.
--
Radoslaw Skorupka
Lodz, Poland
--
BRE Bank SA
ul. Senatorska 18
00-950 Warszawa
www.brebank.pl
Sd Rejonowy dla m. st. Warszawy XII
WydziaGospodarczy Krajowego Rejestru Sdowego, nr
rejestru przedsibiorców KRS 0000025237
NIP: 526-021-50-88
Wedug stanu na dzie01.01.2007 r. kapitazakadowy
BRE Banku SA (w caoci opacony) wynosi
118.064.140 z. W zwizku z realizacjwarunkowego
podwyszenia kapitau zakadowego, na podstawie
uchwaXVI WZ z dnia 21.05.2003 r., kapitazakadowy
BRE Banku SA moe ulec podwyszeniu do kwoty
118.760.528 z. Akcje w podwyszonym kapitale zakadowym bdw caoci opacone.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Doug Fuerst
Consultant
BK Associates
Brooklyn, NY
[phone] (Office)
[phone] (Fax)
[phone] (Cell)
***@bkassociates.net

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Chris Mason]

Doug

You quoted the my previous post and you didn't retain my quoting of Paul's
original post so your post is a bit difficult to follow by itself.

In fact, Paul's original post indicated that a data set which did *not*
exist was kept although not actually created - and so the SPACE issue did
not arise. It was this conundrum for which I made a hesitating attempt to
try to provide a resolution.

If you've only just discovered this "trick" - pointed out to me many years
ago by a guru from the Washington Systems Center - perhaps this example can
show you how it may be used:

//NET PROC
//*
//IEFBR14 EXEC PGM=IEFBR14
//NCPDUMP DD DSN=USER.NCPDUMP,DISP=(MOD,DELETE),
// VOL=REF=SYS1.PARMLIB,SPACE=(TRK,1)
//*
//NET EXEC PGM=ISTINM01,TIME=1440,REGION=2048K
//STEPLIB DD DSN=SYS1.SSPLIB,DISP=SHR
//VTAMLST DD DSN=USER.VTAMLST,DISP=SHR
//VTAMLIB DD DSN=USER.VTAMLIB,DISP=SHR
// DD DSN=SYS1.VTAMLIB,DISP=SHR
//NCPLOAD DD DSN=USER.NCPLOAD,DISP=SHR
//NCPDUMP DD DSN=USER.NCPDUMP,DISP=(NEW,KEEP),
// VOL=REF=SYS1.PARMLIB,SPACE=(TRK,(1,50),RLSE),
// DCB=(RECFM=F,LRECL=512,BLKSIZE=512)

It would appear that all the years I was using this "trick", I may not have
needed the SPACE=(TRK,1) in the first step.

Note that I was obliged to include TIME=1440 on the EXEC statement since I
was obliged to remove SYST from the SCHEDxx entry for ISTINM01.

Chris Mason

----- Original Message -----
From: "Doug Fuerst" <***@ibm-main.lst>
Newsgroups: bit.listserv.ibm-main
To: <IBM-***@BAMA.UA.EDU>
Sent: Wednesday, May 23, 2007 8:19 AM
Subject: Re: DISP=(MOD,DELETE)

Actually, this was recently explained to me by IBM. (MOD,DELETE) indicates
that the dataset might exist, if it does, then MOD it. If it does not,
then create it and then delete it. But SPACE would be required in any case
where the dataset did not exist. This would also be true for (MOD,KEEP).
In this case, if the dataset did not exist it would be created, and KEPT,
but not cataloged. SPACE would be required if the dataset did not exist.
Doug

Post by Chris Mason
Paul
You got my attention with the subject line, DISP=(MOD,DELETE), since I
used to encourage students to consider using this "trick" in order to
immunise their JCL against irritating errors when they had usually started
task procedures relating to networking products and they didn't care
whether or not a data set existed when the task was started. The types of
data sets involved were for traces, logs and dumps.
Typically the started task procedure had an IEFBR14 step where the data
sets were definitely erased if they happened to exist - from, typically of
course, any previous execution - using the DISP=(MOD,DELETE) "trick". The
IEFBR14 step was then followed by the "business" step.[1]
What you show in fact is DISP=(MOD,KEEP).
An explanation - not at all necessarily *the* explanation since it's very
much a guess - for the IEF285I message is that, with DISP=MOD, logically
the data set is considered to exist and thus it can logically be "kept".
However, with, DISP=NEW, the data set is considered *not* to exist - and
you get another error presumably earlier in the processing sequence if it
does exist - and so, before it can be "kept", it must actually be created.
In order to be created, the SPACE parameters are needed.
So *I* understand it. Of course whether or not I understand it correctly
is quite another thing!
Chris Mason

snip undone <<<<<

----- Original Message -----
From: "Paul Gilmartin" <***@ibm-main.lst>
Newsgroups: bit.listserv.ibm-main
To: <IBM-***@BAMA.UA.EDU>
Sent: Wednesday, May 23, 2007 12:05 AM
Subject: DISP=(MOD,DELETE)

//STEP EXEC PGM=IEFBR14
//*
//SYSUT1 DD DSN=user.FOO.BAR.SMPLOG,DISP=(MOD,KEEP),
// VOL=SER=TSO026,
// UNIT=SYSALLDA
where 'user.FOO.BAR.SMPLOG' does not pre-exist on the volume
IEF236I ALLOC. FOR MODKEEP STEP IEF237I
3ECA ALLOCATED TO SYSUT1 IEF142I MODKEEP
STEP - STEP WAS EXECUTED - COND CODE 0000 IEF285I
user.FOO.BAR.SMPLOG KEPT IEF285I VOL SER
NOS= TSO026. IEF373I STEP/STEP
/START 2007142.1554
... but VTOC listing does not show the data set. As I read
the " 12.19.11 Summary of Disposition Processing" in the JCL
RM, I believe it should be kept. When I change to DISP=(NEW,KEEP),
IGD17045I SPACE NOT SPECIFIED FOR ALLOCATION OF DATA SET
Is this working as expected? Should I be able to understand
this?
Thanks,
gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[R.S.]

Gil,
Do the same test for SMS-managed dataset. You'll get JCL error in first case (MOD for non-existent dataset).

OMHO the only thing to understand is: When you use NEW for DASD dataste, you have to specify SPACE. MOD can behave as NEW, the same requirements apply.
--
Radoslaw Skorupka
Lodz, Poland


--
BRE Bank SA
ul. Senatorska 18
00-950 Warszawa
www.brebank.pl

Sd Rejonowy dla m. st. Warszawy
XII Wydzia Gospodarczy Krajowego Rejestru Sdowego,
nr rejestru przedsibiorców KRS 0000025237
NIP: 526-021-50-88
Wedug stanu na dzie 01.01.2007 r. kapita zakadowy BRE Banku SA (w caoci opacony) wynosi 118.064.140 z. W zwizku z realizacj warunkowego podwyszenia kapitau zakadowego, na podstawie uchwa XVI WZ z dnia 21.05.2003 r., kapita zakadowy BRE Banku SA moe ulec podwyszeniu do kwoty 118.760.528 z. Akcje w podwyszonym kapitale zakadowym bd w caoci opacone.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[FRASER, Brian]

I think it should work for an SMS dataset if the DATACLAS assigned by
the ACS has default space attributes.

<snip>

Gil,
Do the same test for SMS-managed dataset. You'll get JCL error in first
case (MOD for non-existent dataset).

OMHO the only thing to understand is: When you use NEW for DASD dataste,
you have to specify SPACE. MOD can behave as NEW, the same requirements
apply.
--
Radoslaw Skorupka
Lodz, Poland

</snip>

-----------------------------------------------------------------------------------
This e-mail is sent by Suncorp-Metway Limited ABN 66 010 831 722 or one of its related entities "Suncorp".

Suncorp may be contacted at Level 18, 36 Wickham Terrace, Brisbane or on 13 11 55 or at suncorp.com.au.

The content of this e-mail is the view of the sender or stated author and does not necessarily reflect the view of Suncorp. The content, including attachments, is a confidential communication between Suncorp and the intended recipient. If you are not the intended recipient, any use, interference with, disclosure or copying of this e-mail, including attachments, is unauthorised and expressly prohibited. If you have received this e-mail in error please contact the sender immediately and delete the e-mail and any attachments from your system.

If this e-mail constitutes a commercial message of a type that you no longer wish to receive please reply to this e-mail by typing Unsubscribe in the subject line.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[TISLER Zaromil]

Paul,

<snip>
The JCL fragment:

//STEP EXEC PGM=IEFBR14
//*
//SYSUT1 DD DSN=user.FOO.BAR.SMPLOG,DISP=(MOD,KEEP),
// VOL=SER=TSO026,
// UNIT=SYSALLDA

where 'user.FOO.BAR.SMPLOG' does not pre-exist on the volume produces
allocation messages:

IEF236I ALLOC. FOR MODKEEP STEP
IEF237I 3ECA ALLOCATED TO SYSUT1
IEF142I MODKEEP STEP - STEP WAS EXECUTED - COND CODE 0000
IEF285I user.FOO.BAR.SMPLOG KEPT
IEF285I VOL SER NOS= TSO026.
IEF373I STEP/STEP /START 2007142.1554

.. but VTOC listing does not show the data set. As I read the " 12.19.11
Summary of Disposition Processing" in the JCL RM, I believe it should be
kept.
<snip>


according to "12.19 DISP Parameter" in the JCL RM (z/OS 1.8):

----------------------------------------
If the system obtains unit and volume information for an OLD, MOD, or SHR
status, the data set is treated as if it exists, whether or not it is
physically on the device.
----------------------------------------

and the "4.6.2.1.8 Disposition Processing of Data Sets that Do Not Exist" in
the JCL UG (z/OS 1.8) says:

----------------------------------------
When VOLUME and UNIT Are Coded: When you code VOLUME and UNIT on the DD
statement, a JCL error will occur if the problem program attempts to open
the data set. Otherwise, the data set disposition depends on the DISP normal
termination disposition:


When the normal termination disposition is KEEP, the job log will show that
the data set was kept.
----------------------------------------

Zaromil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Paul Gilmartin]

Post by Doug Fuerst
I have a headache....

Sorry for the inappropriate subject line. Fingers on autopilot.

[from R.S.]

Post by Doug Fuerst

Post by R.S.
It depends. Try DISP=(MOD,CATLG). It *works* for
non-existent dataset, ...

Whether it depends, depends. I tried it with a non-SMS prefix.
the data set name was catalogued as on the volume, but DSLIST
Info gave me "Data set not found". Silly. If programmer omits
VOLSER, allocation looks for it and creates it if not found; if
programmer supplies VOLSER, allocation trusts him that it exists.
So I have yet no way to create a data set conditionally if I
want to control the VOLSER.

I need to try an experiment with a specific unit address and
omitting VOLSER. Not practical for applications, and a terrible
way to catalog a data set, but I'm curious.

(I Uncatalogued the stump.)

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Dave Kopischke]

Post by Doug Fuerst
Actually, this was recently explained to me by IBM. (MOD,DELETE)
indicates that the dataset might exist, if it does, then MOD it. If
it does not, then create it and then delete it. But SPACE would be
required in any case where the dataset did not exist. This would also
be true for (MOD,KEEP). In this case, if the dataset did not exist it
would be created, and KEPT, but not cataloged. SPACE would be
required if the dataset did not exist.

The SPACE parameter is required, not necessarily real space on a volume. I
use this DD in an IEFBR14 step to clean up possibly pre-existing files of any
kind:

//SYSUT1 DD DSN=Your.Data.Set.Name,
// DISP=(MOD,DELETE,DELETE),
// UNIT=(SYSDA,,DEFER),
// SPACE=(TRK,(0,0),RLSE)

It follows the rules for if the dataset exists, it will try to extend it. If it doesn't
exist, it will create it and delete it with zero space allocated. If it exists on
any device other than disk, it will still delete it since the DEFER parameter
keeps a UNIT from being allocated until the file is actually opened and a BR14
will never actually open the file.

I believe Gil's example will also fail if the file does not exist and he actually
tries to write to it without a SPACE parameter. If the file does not exist and
you don't at least open and close the file, no VTOC record will be written to
the disk.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[shmuel+ (Shmuel Metz , Seymour J.)]

Post by Dave Kopischke
I believe Gil's example will also fail if the file does not exist and
he actually tries to write to it without a SPACE parameter. If the
file does not exist and you don't at least open and close the file,
no VTOC record will be written to the disk.

It doesn't matter whether there is an OPEN; the ALLOC request to
create the DSCB1 is done by Allocation. If his DD statement doesn't
request an ALLOC then the OPEN will fail on an ABEND S213.
--
Shmuel (Seymour J.) Metz, SysProg and JOAT
ISO position; see <http://patriot.net/~shmuel/resume/brief.html>
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Tom Schmidt]

Gil,

Do you have anything specified for SPACE in the active ALLOCxx member in
your PARMLIB?

--
Tom Schmidt
Madison, WI


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Patrick O'Keefe]

On Wed, 23 May 2007 13:33:19 -0500, Dave Kopischke

Post by Dave Kopischke
...
//SYSUT1 DD DSN=Your.Data.Set.Name,
// DISP=(MOD,DELETE,DELETE),
// UNIT=(SYSDA,,DEFER),
// SPACE=(TRK,(0,0),RLSE)
...

Good thing you're releasing all that unused space. :-)
Does RLSE have some value that isn't immediately obvious in this case?

Pat O'Keefe

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Chris Mason]

Pat

It was the second DELETE that caught my attention given that Dave had said
his program was IEFBR14. <g>

Chris Mason

----- Original Message -----
From: "Patrick O'Keefe" <***@ibm-main.lst>
Newsgroups: bit.listserv.ibm-main
To: <IBM-***@BAMA.UA.EDU>
Sent: Wednesday, May 23, 2007 10:07 PM
Subject: Re: DISP=(MOD,DELETE)

Post by Patrick O'Keefe
On Wed, 23 May 2007 13:33:19 -0500, Dave Kopischke

Post by Dave Kopischke
...
//SYSUT1 DD DSN=Your.Data.Set.Name,
// DISP=(MOD,DELETE,DELETE),
// UNIT=(SYSDA,,DEFER),
// SPACE=(TRK,(0,0),RLSE)
...

Good thing you're releasing all that unused space. :-)
Does RLSE have some value that isn't immediately obvious in this case?
Pat O'Keefe

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Dave Kopischke]

Post by Patrick O'Keefe

Post by Dave Kopischke
...
//SYSUT1 DD DSN=Your.Data.Set.Name,
// DISP=(MOD,DELETE,DELETE),
// UNIT=(SYSDA,,DEFER),
// SPACE=(TRK,(0,0),RLSE)
...

Good thing you're releasing all that unused space. :-)
Does RLSE have some value that isn't immediately obvious in this case?

Yes, lot's of value:
1) It's proof positive that I am, in fact, a creature of habit.
2) Implies I am genetically incapable of coding space allocations without a
release since, in this case, it serves no useful purpose. But you gotta admit it
looks good. All lined up with one parameter per line....

I'll think up some really stupid reasons later...

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Farley, Peter x23353]

Actually, SPACE=(0,0) will work just as well, if you can get it past your
SDLC (software development life cycle) software (ours requires
SPACE=(CYL,0), all space allocations must be CYL by fiat).

But you're right, it surely shows you for the creature of habit that you
are... :) Welcome to the club!

Peter

-----Original Message-----
Sent: Wednesday, May 23, 2007 5:04 PM
Subject: Re: DISP=(MOD,DELETE)

Post by Patrick O'Keefe

Post by Dave Kopischke
...
//SYSUT1 DD DSN=Your.Data.Set.Name,
// DISP=(MOD,DELETE,DELETE),
// UNIT=(SYSDA,,DEFER),
// SPACE=(TRK,(0,0),RLSE)
...

Good thing you're releasing all that unused space. :-)
Does RLSE have some value that isn't immediately obvious in this case?

1) It's proof positive that I am, in fact, a creature of habit.
2) Implies I am genetically incapable of coding space allocations without
a release since, in this case, it serves no useful purpose. But you gotta
admit it looks good. All lined up with one parameter per line....
I'll think up some really stupid reasons later...

This message and any attachments are intended only for the use of the addressee and
may contain information that is privileged and confidential. If the reader of the
message is not the intended recipient or an authorized representative of the
intended recipient, you are hereby notified that any dissemination of this
communication is strictly prohibited. If you have received this communication in
error, please notify us immediately by e-mail and delete the message and any
attachments from your system.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Paul Gilmartin]

Post by Farley, Peter x23353
Actually, SPACE=(0,0) will work just as well, if you can get it past your
SDLC (software development life cycle) software (ours requires
SPACE=(CYL,0), all space allocations must be CYL by fiat).

Will they permit SPACE=(0,(0,0),,CONTIG,ROUND)?

Will allocating 0 cylinders succeed if the volume has no
free cylinders?

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Farley, Peter x23353]

-----Original Message-----
Sent: Wednesday, May 23, 2007 5:32 PM
Subject: Re: DISP=(MOD,DELETE)

Post by Farley, Peter x23353
Actually, SPACE=(0,0) will work just as well, if you can get it past your
SDLC (software development life cycle) software (ours requires
SPACE=(CYL,0), all space allocations must be CYL by fiat).

Will they permit SPACE=(0,(0,0),,CONTIG,ROUND)?

No. 1st arg of SPACE must be CYL, no exceptions. Like I said, it's by fiat,
controlled by JCL scanning software.

Will allocating 0 cylinders succeed if the volume has no
free cylinders?

No clue, never had that situation come up. Since SMS and storage admins, we
simple programmers can't play games as we once did, like filling up a
particular scratch disk just to see what happens if...

This message and any attachments are intended only for the use of the addressee and
may contain information that is privileged and confidential. If the reader of the
message is not the intended recipient or an authorized representative of the
intended recipient, you are hereby notified that any dissemination of this
communication is strictly prohibited. If you have received this communication in
error, please notify us immediately by e-mail and delete the message and any
attachments from your system.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Bob Rutledge]

Never come up here either, but I have a litterbox.

Z890 e . : ZOSPL1
Command ===>

Unit . . : 3390

Volume Data VTOC Data Free Space Tracks Cyls
Tracks . : 50,085 Tracks . : 14 Size . . : 0 0
%Used . : 100 %Used . . : 1 Largest . : 0 0
Trks/Cyls: 15 Free DSCBS: 696 Free
Extents . : 0

//S010 EXEC PGM=IEFBR14

//DD1 DD DISP=(,PASS),DSN=&&NOCYL,UNIT=SYSDA,

// VOL=SER=ZOSPL1,SPACE=(CYL,(0,0),RLSE)


IEF236I ALLOC. FOR TS0001BR S010

IGD100I 2381 ALLOCATED TO DDNAME DD1 DATACLAS ( )

IEF142I TS0001BR S010 - STEP WAS EXECUTED - COND CODE 0000


IEF237I 2381 ALLOCATED TO SYS00001

IEF285I SYS07143.T174703.RA000.TS0001BR.R0106310 KEPT

IEF285I VOL SER NOS= ZOSPL1.

IEF285I SYS07143.T174703.RA000.TS0001BR.NOCYL.H01 DELETED

IEF285I VOL SER NOS= ZOSPL1.


Bob

Post by Farley, Peter x23353

-----Original Message-----
Sent: Wednesday, May 23, 2007 5:32 PM
Subject: Re: DISP=(MOD,DELETE)

Post by Farley, Peter x23353
Actually, SPACE=(0,0) will work just as well, if you can get it past your
SDLC (software development life cycle) software (ours requires
SPACE=(CYL,0), all space allocations must be CYL by fiat).

Will they permit SPACE=(0,(0,0),,CONTIG,ROUND)?

No. 1st arg of SPACE must be CYL, no exceptions. Like I said, it's by fiat,
controlled by JCL scanning software.

Will allocating 0 cylinders succeed if the volume has no
free cylinders?

No clue, never had that situation come up. Since SMS and storage admins, we
simple programmers can't play games as we once did, like filling up a
particular scratch disk just to see what happens if...

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Ted MacNEIL]

Post by Farley, Peter x23353
all space allocations must be CYL by fiat

Since ECKD came out with DEFINE EXTENT that doesn't matter anymore!

The control unit no longer signals back asking if it's okay to move to the next in line (cylinder or track).
And, even if it did, it's nano's.

-
Too busy driving to stop for gas!

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Ted MacNEIL]

Will allocating 0 cylinders succeed if the volume has no free cylinders?

Actually, yes.
As long as your VTOC has room.

Remember setting GDG models?

-
Too busy driving to stop for gas!

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Dave Kopischke]

Post by Paul Gilmartin
Will they permit SPACE=(0,(0,0),,CONTIG,ROUND)?
Will allocating 0 cylinders succeed if the volume has no
free cylinders?

Do you really want to know this stuff ???? Or are you just trying to send
someones brain into an infinite loop around the longest string of obsolete
parameters you can think of ????

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Paul Gilmartin]

Post by Dave Kopischke

Post by Paul Gilmartin
Will they permit SPACE=(0,(0,0),,CONTIG,ROUND)?

That was just poking fun at Peter's plaint about the totalitarian
administration. And I didn't even ask whether they also require
directories to be allocated in multiples of 45 blocks - 1.

Post by Dave Kopischke

Post by Paul Gilmartin
Will allocating 0 cylinders succeed if the volume has no
free cylinders?

Do you really want to know this stuff ???? Or are you just trying to send
someones brain into an infinite loop around the longest string of obsolete
parameters you can think of ????

Sort of an early Friday topic. But someone answered, "Yes, it does
work." I like that (something (minor) I like about z/OS?) I like
it when designers get the boundary conditions right. And I deem
it particularly elegant if they don't need to handle zero as a
special case. And I can't mentally code this one so that zero is
handled properly without special casing, in the mainline code.

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Paul Gilmartin]

Post by Chris Mason
It was the second DELETE that caught my attention given that Dave had said
his program was IEFBR14. <g>

Well, the operator might decide to cancel it.

Post by Chris Mason

Post by Dave Kopischke
//SYSUT1 DD DSN=Your.Data.Set.Name,
// DISP=(MOD,DELETE,DELETE),
...

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Paul Gilmartin]

Post by Chris Mason
If you've only just discovered this "trick" - pointed out to me many years
ago by a guru from the Washington Systems Center - perhaps this example can
//NET PROC
//*
//IEFBR14 EXEC PGM=IEFBR14
//NCPDUMP DD DSN=USER.NCPDUMP,DISP=(MOD,DELETE),
// VOL=REF=SYS1.PARMLIB,SPACE=(TRK,1)
//*
//NET EXEC PGM=ISTINM01,TIME=1440,REGION=2048K
//STEPLIB DD DSN=SYS1.SSPLIB,DISP=SHR
//VTAMLST DD DSN=USER.VTAMLST,DISP=SHR
//VTAMLIB DD DSN=USER.VTAMLIB,DISP=SHR
// DD DSN=SYS1.VTAMLIB,DISP=SHR
//NCPLOAD DD DSN=USER.NCPLOAD,DISP=SHR
//NCPDUMP DD DSN=USER.NCPDUMP,DISP=(NEW,KEEP),
// VOL=REF=SYS1.PARMLIB,SPACE=(TRK,(1,50),RLSE),
// DCB=(RECFM=F,LRECL=512,BLKSIZE=512)

Would it work equally well with VOL=SER as VOL=REF? (I suspect that
was for portability. But I'd expect my admins to be pretty fussy
about who creates what on the PARMLIB volume.)

Well, my objectives were (arising from the ISV SMP/E "gripes" thread):

o Create the data set if it does not exist.

o Append to (not replace) the data set if it does exist.

o Allow the user optionally to specify VOLSER.

On experimenting, reading, and receiving input from this list,
I've lowered my sights and retreated to DISP=(NEW,{KEEP|CATLG})

Post by Chris Mason
Note that I was obliged to include TIME=1440 on the EXEC statement since I
was obliged to remove SYST from the SCHEDxx entry for ISTINM01.

I don't understand that part at all.

Thanks for the input,
gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Chris Mason]

Paul

I provided my example because your subject title pulled me it,
DISP=(MOD,DELETE) being the heart - or so I thought until this thread
unravelled - of the "trick" to starting with a new data set in a following
job step. Thus the objectives of this technique and what you are after
differ.

Regarding SYST: I did provide a reference in an earlier post to an earlier
thread where I solicited comments on the matter of removing SYST. However if
you Google on "SYST SCHEDxx Mason", you'll get one hit, the URL of the
thread in an archive of IBM-MAIN.

Regarding VOL=REF (or VOL=SER): My example is very close to what I used to
use - only the names have been changed to make them more general. Very
probably - almost certainly - you can replace VOL=REF with to whatever it is
normally equivalent.

Regarding SYS1.PARMLIB: From time to time when posting I explain that my
background is *not* production but education/test. I used to run up to 8 MVS
systems where I had one volume per system, the "read/write" volume, and one
common "read-only" volume. I suspect that the SYS1.PARMLIB data set was one
the "read/write" volume for some reason I have now forgotten.

Chris Mason

----- Original Message -----
From: "Paul Gilmartin" <***@ibm-main.lst>
Newsgroups: bit.listserv.ibm-main
To: <IBM-***@BAMA.UA.EDU>
Sent: Thursday, May 24, 2007 5:05 AM
Subject: Re: DISP=(MOD,DELETE)

Post by Paul Gilmartin

Post by Chris Mason
If you've only just discovered this "trick" - pointed out to me many years
ago by a guru from the Washington Systems Center - perhaps this example can
//NET PROC
//*
//IEFBR14 EXEC PGM=IEFBR14
//NCPDUMP DD DSN=USER.NCPDUMP,DISP=(MOD,DELETE),
// VOL=REF=SYS1.PARMLIB,SPACE=(TRK,1)
//*
//NET EXEC PGM=ISTINM01,TIME=1440,REGION=2048K
//STEPLIB DD DSN=SYS1.SSPLIB,DISP=SHR
//VTAMLST DD DSN=USER.VTAMLST,DISP=SHR
//VTAMLIB DD DSN=USER.VTAMLIB,DISP=SHR
// DD DSN=SYS1.VTAMLIB,DISP=SHR
//NCPLOAD DD DSN=USER.NCPLOAD,DISP=SHR
//NCPDUMP DD DSN=USER.NCPDUMP,DISP=(NEW,KEEP),
// VOL=REF=SYS1.PARMLIB,SPACE=(TRK,(1,50),RLSE),
// DCB=(RECFM=F,LRECL=512,BLKSIZE=512)

Would it work equally well with VOL=SER as VOL=REF? (I suspect that
was for portability. But I'd expect my admins to be pretty fussy
about who creates what on the PARMLIB volume.)
o Create the data set if it does not exist.
o Append to (not replace) the data set if it does exist.
o Allow the user optionally to specify VOLSER.
On experimenting, reading, and receiving input from this list,
I've lowered my sights and retreated to DISP=(NEW,{KEEP|CATLG})

Post by Chris Mason
Note that I was obliged to include TIME=1440 on the EXEC statement since I
was obliged to remove SYST from the SCHEDxx entry for ISTINM01.

I don't understand that part at all.
Thanks for the input,
gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Brian Peterson]

The attribute SYST from SCHEDxx is only applicable to started tasks with one
job step. Chris had added an IEFBR14 step to his NET proc, thus VTAM starts
with the following message:

IEF188I PROBLEM PROGRAM ATTRIBUTES ASSIGNED

This message signifies that the SYST attribute was not honored. Thus, he
had to code TIME=1440 to achieve the same result.

Brian
(snip)

Post by Paul Gilmartin

Post by Chris Mason
Note that I was obliged to include TIME=1440 on the EXEC statement since I
was obliged to remove SYST from the SCHEDxx entry for ISTINM01.

I don't understand that part at all.
Thanks for the input,
gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[Dave Kopischke]

Post by Paul Gilmartin
On experimenting, reading, and receiving input from this list,
I've lowered my sights and retreated to DISP=(NEW,{KEEP|CATLG})

Be careful with your disposition on good completion with MOD. I came across
an issue years ago where JCL made it into production with MOD,KEEP,KEEP.
the dataset went multi-volume and since the second parameter wasn't CATLG,
the dataset just ended at the first volume. I had to manually track down the
second piece and manually glue it back together. Very strange, but totally
understandable.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

[shmuel+ (Shmuel Metz , Seymour J.)]

... but VTOC listing does not show the data set. As I read the "
12.19.11 Summary of Disposition Processing" in the JCL RM, I believe
it should be kept.

There was nothing to keep. You provided a volume serial, thus
bypassing the special Allocation processing of MOD.

Is this working as expected?

It's working as documented.
--
Shmuel (Seymour J.) Metz, SysProg and JOAT
ISO position; see <http://patriot.net/~shmuel/resume/brief.html>
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html