Strange TEMPLATE / LISTDEF Problem with COPY

Cuneyt Goksu

Strange TEMPLATE / LISTDEF Problem with COPY
Hi Members,

Here is DB2 V7, OS/390 R10, Tape Management System RMM and 3480 Tape Hardware.


I have a strange problem with TEMPLATEs and COPY.

Below is the SYSIN of the IC JCL.

-----------
TEMPLATE TEMP_1
UNIT CART
DSN 'DB2TCOPY.&TS.(+1)'
GDGLIMIT 10
DISP (NEW,CATLG,DELETE)
STACK YES
MODELDCB 'MODEL.DSCB'
TRTCH COMP
RETPD 7

LISTDEF LIST_1 INCLUDE TABLESPACES DATABASE DFBS

COPY LIST LIST_1 COPYDDN TEMP_1 SHRLEVEL CHANGE
-----------

It requests tape mount and receives scratch tape 2097. When it's full it
requests second tape 2084. Then it requests the first tape (2097) again.

It basically loops!
Finally I cancel the job.

Below there's a sample output.

There is no DB2 specific code in the output.

This JCL template and model worked for many other shops but not here.

Do you have any comments for that?

Regards,
Cuneyt

--------------
ICH70001I SYSADM LAST ACCESS AT 09:16:56 ON TUESDAY, NOVEMBER 30, 2004
IHASP373 DBTDAY1 STARTED - INIT 1 - CLASS A - SYS TEST
IEF403I DBTDAY1 - STARTED - TIME=09.26.55

*IEF233D M 0C68,PRIVAT,SL,DBTDAY1,S1,DB2TCOPY.ADBSCC.G0002V00,
OR RESPOND TO IEF455D MESSAGE
*64 IEF455D MOUNT PRIVAT ON 0C68 FOR DBTDAY1 S1 OR REPLY 'NO'
IEC705I TAPE ON 0C68,002097,SL,COMP,DBTDAY1,S1,DB2TCOPY.ADBSCC.G0002V00
IEC502E K 0C68,002097,SL,DBTDAY1,S1,DB2TCOPY.SAFBS03C.G0191V00
*IEC501A M 0C68,PRIVAT,SL,COMP,DBTDAY1,S1,DB2TCOPY.SAFBS03C.G0191V00
EDG4027I VOLUME 001881 REJECTED. IT IS NOT A SCRATCH VOLUME AND MOUNT REQUEST

EDG4005E VOLUME 001881 ON 0C68 REJECTED FOR USE BY DBTDAY1, S1, SYS00001
IEC502E K 0C68,001881,SL,DBTDAY1,S1,DB2TCOPY.SAFBS03C.G0191V00
*IEC501A M 0C68,PRIVAT,SL,COMP,DBTDAY1,S1,DB2TCOPY.SAFBS03C.G0191V00
IEC705I TAPE ON 0C68,002084,SL,COMP,DBTDAY1,S1,DB2TCOPY.SAFBS03C.G0191V00
IEC502E K 0C68,002084,SL,DBTDAY1,S1
*IEC501A M 0C68,002097,SL,COMP,DBTDAY1,S1,DB2TCOPY.SAFBS03D.G0191V00
IEF450I DBTDAY1 S1 - ABEND=S222 U0000 REASON=00000000 TIME=09.33.14
IEF234E K 0C68,002097,PVT,DBTDAY1,S1
-----------

---------------------------------------------------------------------------------
Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at [login to unmask email] Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm

Kornelis Abe

Re: Strange TEMPLATE / LISTDEF Problem with COPY
(in response to Cuneyt Goksu)
Cuneyt,

your problem might be related to a PMR I recently opened with IBM.
It is PMR 89363,211,788 from which I have taken the following
excerpts:

<PMR>
Problem:
During catalog reorg using listdef and template processing we
receive repeatedly the following message:
IEC507D E 1314,W40792,DB1DREO,REORG.REORG,4322.G091012.U00042 IEC507D REPLY 'U'-USE OR 'M'-UNLOAD

It seems the job is trying to overwrite a dataset created in the
same job step. Also it appears the DSN in the message is truncated,
giving only the 'remainder' of the data set name.

The problem appears to be with the unload datasets, which are
written to (virtual) tape with the STACK YES option:
TEMPLATE UNLDDATA
DSN 'DB1D.&DB..&SN..D&JU(3,5)..G&TI..U&PA(3,3).'
DISP (NEW,DELETE,CATLG)
UNIT VSM01
RETPD 7
VOLCNT 99
STACK YES -- FOR TAPE ALLOCATIONS --
</PMR>

After some discussion with the lab in the course of problem analysis
I wrote:
<PMR>
It looks like this problem breeaks down into
three separate DB2 problems:

1) Unload dataset names are generated erroneously: the reorg
processes SYSCOPY, SYSDBASE, SYSDBAUT, then SYSDDF. The unload
dataset names reported on the reorg's sysprint are ok, however the
unload dataset names in JES2's JESMSGLG use SYSCOPY, SYSCOPY,
SYSCOPY, then SYSDDF.
The reuse of SYSCOPY causes duplicate DSNames to be generated.

2) The time stamps used for DSN generation are not current
timestamps, rather they are start-of-jobstep timestamps. Again,
this causes duplicate DSNames to be generated.

3) Datasets are not stacked on the tape, rather the tape is rewound
every time. I can only guess this is due to the reuse of dataset
names. It also happens with

Closer inspection and comparison of REORG output with JES2 msglog
reveals two additional problems:

4) inline full-image copy dataset of SYSDBASE and SYSDBAUT do not
appear in the JES log, even though they appear correctly in the
REORG's sysprint.

5) File sequence numbers as reported by reorg sysprint do not match
the file sequence numbers that JES2 reports.

Fortunately, the data in SYSIBM.SYSCOPY seems to be correct. The
inline FIC datasets produced by REORG are recorded only when not
skipped. I.e. (e.g.) SYSDDF is present, but SYSDBASE and SYSDBAUT
are missing from SYSCOPY. SYSCOPY itself is missing of course...
</PMR>

The lab seems to acknowledge something's wrong here.
They are actively investigating the problem.

Hope this helps.
Abe.
===

-----Original Message-----
From: DB2 Data Base Discussion List [mailto:[login to unmask email]On
Behalf Of Cuneyt GOKSU
Sent: dinsdag 30 november 2004 10:31
To: [login to unmask email]
Subject: Strange TEMPLATE / LISTDEF Problem with COPY


Hi Members,

Here is DB2 V7, OS/390 R10, Tape Management System RMM and 3480 Tape Hardware.


I have a strange problem with TEMPLATEs and COPY.

COPY LIST LIST_1 COPYDDN TEMP_1 SHRLEVEL CHANGE

It requests tape mount and receives scratch tape 2097. When it's full it
requests second tape 2084. Then it requests the first tape (2097) again.

It basically loops!
Finally I cancel the job.

---------------------------------------------------------------------------------
Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at [login to unmask email] Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm

Graeme St.Clair

Re: Strange TEMPLATE / LISTDEF Problem with COPY
(in response to Kornelis Abe)
Just a short comment. Unless things have changed since I used to be a
sysprog (not impossible!), only the last 17 characters of the DSN are
recorded on an actual SL tape. See
http://publibfp.boulder.ibm.com/cgi-bin/bookmgr/BOOKS/igg3m300/CCONTENTS
para 2.14
for the official scoop (as it was in 1991!).

Rgds, G.

-----Original Message-----
From: DB2 Data Base Discussion List [mailto:[login to unmask email] On Behalf
Of Kornelis Abe
Sent: Tuesday, November 30, 2004 5:23 AM
To: [login to unmask email]
Subject: Re: [DB2-L] Strange TEMPLATE / LISTDEF Problem with COPY

Cuneyt,

your problem might be related to a PMR I recently opened with IBM.
It is PMR 89363,211,788 from which I have taken the following
excerpts:

<PMR>
Problem:
During catalog reorg using listdef and template processing we receive
repeatedly the following message:
IEC507D E 1314,W40792,DB1DREO,REORG.REORG,4322.G091012.U00042 IEC507D REPLY
'U'-USE OR 'M'-UNLOAD

It seems the job is trying to overwrite a dataset created in the
same job step. Also it appears the DSN in the message is truncated,
giving only the 'remainder' of the data set name.

The problem appears to be with the unload datasets, which are
written to (virtual) tape with the STACK YES option:
TEMPLATE UNLDDATA
DSN 'DB1D.&DB..&SN..D&JU(3,5)..G&TI..U&PA(3,3).'
DISP (NEW,DELETE,CATLG)
UNIT VSM01
RETPD 7
VOLCNT 99
STACK YES -- FOR TAPE ALLOCATIONS --
</PMR>

After some discussion with the lab in the course of problem analysis I
wrote:
<PMR>
It looks like this problem breeaks down into
three separate DB2 problems:

1) Unload dataset names are generated erroneously: the reorg
processes SYSCOPY, SYSDBASE, SYSDBAUT, then SYSDDF. The unload
dataset names reported on the reorg's sysprint are ok, however the
unload dataset names in JES2's JESMSGLG use SYSCOPY, SYSCOPY,
SYSCOPY, then SYSDDF.
The reuse of SYSCOPY causes duplicate DSNames to be generated.

2) The time stamps used for DSN generation are not current
timestamps, rather they are start-of-jobstep timestamps. Again,
this causes duplicate DSNames to be generated.

3) Datasets are not stacked on the tape, rather the tape is rewound
every time. I can only guess this is due to the reuse of dataset
names. It also happens with

Closer inspection and comparison of REORG output with JES2 msglog
reveals two additional problems:

4) inline full-image copy dataset of SYSDBASE and SYSDBAUT do not
appear in the JES log, even though they appear correctly in the
REORG's sysprint.

5) File sequence numbers as reported by reorg sysprint do not match
the file sequence numbers that JES2 reports.

Fortunately, the data in SYSIBM.SYSCOPY seems to be correct. The
inline FIC datasets produced by REORG are recorded only when not
skipped. I.e. (e.g.) SYSDDF is present, but SYSDBASE and SYSDBAUT
are missing from SYSCOPY. SYSCOPY itself is missing of course...
</PMR>

The lab seems to acknowledge something's wrong here.
They are actively investigating the problem.

Hope this helps.
Abe.
===

-----Original Message-----
From: DB2 Data Base Discussion List [mailto:[login to unmask email]On Behalf Of
Cuneyt GOKSU
Sent: dinsdag 30 november 2004 10:31
To: [login to unmask email]
Subject: Strange TEMPLATE / LISTDEF Problem with COPY


Hi Members,

Here is DB2 V7, OS/390 R10, Tape Management System RMM and 3480 Tape
Hardware.


I have a strange problem with TEMPLATEs and COPY.

COPY LIST LIST_1 COPYDDN TEMP_1 SHRLEVEL CHANGE

It requests tape mount and receives scratch tape 2097. When it's full it
requests second tape 2084. Then it requests the first tape (2097) again.

It basically loops!
Finally I cancel the job.

----------------------------------------------------------------------------
-----
Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home
page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select
"Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org.
The IDUG List Admins can be reached at [login to unmask email] Find
out the latest on IDUG conferences at http://conferences.idug.org/index.cfm

---------------------------------------------------------------------------------
Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at [login to unmask email] Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm

Kornelis Abe

Re: Strange TEMPLATE / LISTDEF Problem with COPY
(in response to Graeme St.Clair)
Graeme,

You are entirely correct. I didn't know, so that part of the
problem was easily solved. The other issues are still under
IBM scrutiny.

Abe.
===

-----Original Message-----
From: DB2 Data Base Discussion List [mailto:[login to unmask email]On
Behalf Of Graeme St. Clair

Just a short comment. Unless things have changed since I used to be a
sysprog (not impossible!), only the last 17 characters of the DSN are
recorded on an actual SL tape. See
http://publibfp.boulder.ibm.com/cgi-bin/bookmgr/BOOKS/igg3m300/CCONTENTS
para 2.14
for the official scoop (as it was in 1991!).

Rgds, G.

---------------------------------------------------------------------------------
Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at [login to unmask email] Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm

Cuneyt Goksu

Re: Strange TEMPLATE / LISTDEF Problem with COPY
(in response to Kornelis Abe)
Abe,

Thanks for it.

I informed our IBM rep. about that PMR. He'll also open one for my case.

Cuneyt

---
Cuneyt,

your problem might be related to a PMR I recently opened with IBM.
It is PMR 89363,211,788 from which I have taken the following=20
excerpts:

----

---------------------------------------------------------------------------------
Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at [login to unmask email] Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm

Jim McAlpine

Re: Strange TEMPLATE / LISTDEF Problem with COPY
(in response to Cuneyt Goksu)
No, they haven't changed. The tape header labels only contain the last 17
characters of the dataset name from when the tape was written.

Jim McAlpine

-----Original Message-----
From: Graeme St. Clair [mailto:[login to unmask email]
Sent: 01 December 2004 00:01
To: [login to unmask email]
Subject: Re: Strange TEMPLATE / LISTDEF Problem with COPY


Just a short comment. Unless things have changed since I used to be a
sysprog (not impossible!), only the last 17 characters of the DSN are
recorded on an actual SL tape. See
http://publibfp.boulder.ibm.com/cgi-bin/bookmgr/BOOKS/igg3m300/CCONTENTS
para 2.14
for the official scoop (as it was in 1991!).

Rgds, G.

-----Original Message-----
From: DB2 Data Base Discussion List [mailto:[login to unmask email] On Behalf
Of Kornelis Abe
Sent: Tuesday, November 30, 2004 5:23 AM
To: [login to unmask email]
Subject: Re: [DB2-L] Strange TEMPLATE / LISTDEF Problem with COPY

Cuneyt,

your problem might be related to a PMR I recently opened with IBM.
It is PMR 89363,211,788 from which I have taken the following
excerpts:

<PMR>
Problem:
During catalog reorg using listdef and template processing we receive
repeatedly the following message:
IEC507D E 1314,W40792,DB1DREO,REORG.REORG,4322.G091012.U00042 IEC507D REPLY
'U'-USE OR 'M'-UNLOAD

It seems the job is trying to overwrite a dataset created in the
same job step. Also it appears the DSN in the message is truncated,
giving only the 'remainder' of the data set name.

The problem appears to be with the unload datasets, which are
written to (virtual) tape with the STACK YES option:
TEMPLATE UNLDDATA
DSN 'DB1D.&DB..&SN..D&JU(3,5)..G&TI..U&PA(3,3).'
DISP (NEW,DELETE,CATLG)
UNIT VSM01
RETPD 7
VOLCNT 99
STACK YES -- FOR TAPE ALLOCATIONS --
</PMR>

After some discussion with the lab in the course of problem analysis I
wrote:
<PMR>
It looks like this problem breeaks down into
three separate DB2 problems:

1) Unload dataset names are generated erroneously: the reorg
processes SYSCOPY, SYSDBASE, SYSDBAUT, then SYSDDF. The unload
dataset names reported on the reorg's sysprint are ok, however the
unload dataset names in JES2's JESMSGLG use SYSCOPY, SYSCOPY,
SYSCOPY, then SYSDDF.
The reuse of SYSCOPY causes duplicate DSNames to be generated.

2) The time stamps used for DSN generation are not current
timestamps, rather they are start-of-jobstep timestamps. Again,
this causes duplicate DSNames to be generated.

3) Datasets are not stacked on the tape, rather the tape is rewound
every time. I can only guess this is due to the reuse of dataset
names. It also happens with

Closer inspection and comparison of REORG output with JES2 msglog
reveals two additional problems:

4) inline full-image copy dataset of SYSDBASE and SYSDBAUT do not
appear in the JES log, even though they appear correctly in the
REORG's sysprint.

5) File sequence numbers as reported by reorg sysprint do not match
the file sequence numbers that JES2 reports.

Fortunately, the data in SYSIBM.SYSCOPY seems to be correct. The
inline FIC datasets produced by REORG are recorded only when not
skipped. I.e. (e.g.) SYSDDF is present, but SYSDBASE and SYSDBAUT
are missing from SYSCOPY. SYSCOPY itself is missing of course...
</PMR>

The lab seems to acknowledge something's wrong here.
They are actively investigating the problem.

Hope this helps.
Abe.
===

-----Original Message-----
From: DB2 Data Base Discussion List [mailto:[login to unmask email]On Behalf Of
Cuneyt GOKSU
Sent: dinsdag 30 november 2004 10:31
To: [login to unmask email]
Subject: Strange TEMPLATE / LISTDEF Problem with COPY


Hi Members,

Here is DB2 V7, OS/390 R10, Tape Management System RMM and 3480 Tape
Hardware.


I have a strange problem with TEMPLATEs and COPY.

COPY LIST LIST_1 COPYDDN TEMP_1 SHRLEVEL CHANGE

It requests tape mount and receives scratch tape 2097. When it's full it
requests second tape 2084. Then it requests the first tape (2097) again.

It basically loops!
Finally I cancel the job.

----------------------------------------------------------------------------
-----
Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home
page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select
"Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org.
The IDUG List Admins can be reached at [login to unmask email] Find
out the latest on IDUG conferences at http://conferences.idug.org/index.cfm

----------------------------------------------------------------------------
-----
Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home
page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select
"Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org.
The IDUG List Admins can be reached at [login to unmask email] Find
out the latest on IDUG conferences at http://conferences.idug.org/index.cfm


This email and any files transmitted with it are confidential
and intended solely for the use of the individual or entity to
whom they are addressed. Any unauthorised distribution or
copying is strictly prohibited. Whilst Cedar Software Ltd
takes steps to prevent the transmission of viruses via e-mail,
we cannot guarantee that any email or attachment is free from
computer viruses and you are strongly advised to undertake your
own anti-virus precautions. Cedar Software Ltd grants no
warranties regarding performance, use or quality of any e-mail
or attachment and undertakes no liability for loss or damage,
howsoever caused.


For more information on Cedar Software and our products,
please visit our web site at http://www.cedar.com/Home.asp

---------------------------------------------------------------------------------
Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at [login to unmask email] Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm

James Campbell

Re: Strange TEMPLATE / LISTDEF Problem with COPY
(in response to Jim McAlpine)
As a comment, it is usually considered good form to do the IC of
SYSCOPY after the others.

James Campbell

On 30 Nov 2004 at 11:23, Kornelis Abe wrote:

<snip>
>
> 1) Unload dataset names are generated erroneously: the reorg
> processes SYSCOPY, SYSDBASE, SYSDBAUT, then SYSDDF. The unload
> dataset names reported on the reorg's sysprint are ok, however the
> unload dataset names in JES2's JESMSGLG use SYSCOPY, SYSCOPY,
> SYSCOPY, then SYSDDF.
> The reuse of SYSCOPY causes duplicate DSNames to be generated.
>
<rest snipped>

---------------------------------------------------------------------------------
Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at [login to unmask email] Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm