Thanks Michael for this. Interesting to see that this could be DDR's problem 
not to handle EOT.
Maybe I should make a PMR to see what the IBM guys think.  

-----Original Message-----
From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf 
Of Michael Coffin
Sent: 30. marraskuuta 2010 17:55
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: DDR multivolume tape as output

The inability of DDR to handle EOT conditions with anything other than ALTTAPE 
has always been a problem for me too.  Among other things, it ties up TWO 
physical tape drives when only ONE is ever active at any given time (not a 
problem in large shops, HUGE problem in small shops that only have 4 physical 
drives).  Back in 2003/2004 I wrote a mod to HCPDDR and saved the modified 
module as MCIDDR.  Rather than DDR just sitting there waiting for the "next 
tape" when it encounters an EOT condition, I have it SVC202 out to a local exit 
(MCIDDRNT EXEC in the following example, where NT stands for Next Tape!) that 
can be easily customized to do whatever needs to be done.
In my case, I use SL tapes so my MCIDDRNT EXEC would use VMTAPE to mount and 
catalog a scratch tape.  VMTape is configured to interface with our 3494 
robotics which performs the actual MOUNT.  Once VMTAPE delivers the mounted 
tape to our virtual machine, I skip past the SL and leave the tape positioned 
for the next write.  It does other things too, like set a flag that we spanned 
a tape that can be collected by the caller it receives control back from DDR, 
etc.

Definitely use COMPACT on your OUTPUT statement. :)

FYI, I always had trouble restoring the pack that spanned two tapes when using 
SL tapes - so I ended up changing the process to just re-dump the pack that 
spanned the tapes in its entirety as the first spindle on the next tape.  I 
only mention this because, as is with ANY mission-critical function, you should 
absolutely TEST TEST and TEST AGAIN before deciding that your process is 
working (backing up data is great, but if you can't recover it you'll quickly 
discover how useless the backup is.....).  

The HCPDDR mod is only a couple of lines so I'll just attach it below, and I 
see by my comments this based on work originally by Peter Webb at Toronto 
Transit so THANK YOU Peter.  I haven't applied this mod to HCPDDR source since 
2004 (source dated 2/11/2003), so you'll probably need to resequence it. :)

-Mike

./ I 56317202          $ 56317502 300                 06/22/04 08:10:14

* MC CONSULTING: START LOCAL DDR MODIFICATION   PART 1 OF 2
56317502
* SUPPORTED BY:  michaelcof...@deleted.spam
56317802
* MCIDDR IS AN EXACT COPY OF HCPDDR, BUT WE'VE ADDED SUPPORT FOR EOT
56318102
* PROCESSING TO ALLOW VM:TAPE TO MOUNT A SCRATCH TAPE AND CONTINUE.
56318402
* THANKS TO PETER WEBB AT TORONTO TRANSIT SYSTEM FOR ORIGINAL IDEA.
56318702
*
56319002
* THE TWO STATEMENTS THAT PRECEDE THIS MODIFICATION ARE:
56319302
* CKDEOV   DS    0H
56319602
*          BAL   R14,MSG005          GO PRINT THE END OF VOL MS
56319902
         TM    DDRCMSF,DDRCMS      RUNNING UNDER CMS?
56320202
         BZ    MCC1                NO, THEN DO NOT CALL MCCCI EOT EXIT
56320502
         CMSCALL PLIST=MCCEOT,ERROR=*  CALL MCCCI EOT TAPE HANDLER
56320802
MCC1     DS    0H
56321102
* MC CONSULTING: END   LOCAL DDR MODIFICATION   PART 1 OF 2
56321402
*
56321702
./ I 58447802          $ 58448202 400                 06/22/04 08:10:14

*
58448202
* MC CONSULTING: START LOCAL DDR MODIFICATION   PART 2 OF 2
58448602
* SUPPORTED BY:  michaelcof...@deleted.spam
58449002
* THE TWO STATEMENTS THAT PRECEDE THIS MODIFICATION ARE:
58449402
* SIORET7  DS    0H
58449802
*          BRANCH SIOEP,SIORET
58450202
MCCEOT   DC    CL8'EXEC'           POINT TO EXEC INTERPRETER
58450602
         DC    CL8'MCIDDRNT'       NAME OF MCCCI EOT HANDLER.
58451002
         DC    8X'FF'              END OF MCIXTAPE EOT PARAMETER LIST
58451402
* MC CONSULTING: END   LOCAL DDR MODIFICATION   PART 2 OF 2
58451802




-----Original Message-----
From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf 
Of Janne Jarvinen
Sent: Tuesday, November 30, 2010 3:36 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: DDR multivolume tape as output

Hello,

I'm a z/OS person and have some diffuculties to uderstand how to use VTS tapes 
in z/VM environment, hopefully get some help from this forum.
  
How to mount the second DDR dump-tape under VM? 

Want to DUMP my SYSRES to tape:
 
1. MAINT: DFSMSRM MOUNT CAT SCRATCH0 

                                          
DGTUIR2026I DFSMS request 84 accepted for 
processing                            
Ready; T=0.01/0.01 
10:24:22                                                     
TAPE 0D3F ATTACHED TO MAINT 
0D3F                                                
 10:24:24  * MSG FROM RMSMASTR:    FSMBCR2120I Request 84 complete; volume 
V900
 02, category SCRATCH0, mounted on device 0D3F in library 
IBMVTS1               
 10:24:24  * MSG FROM RMSMASTR:    FSMBEC2125I Request 84: device 0D3F 
attached
  to MAINT as 0D3F; access mode = 
READWRITE                                     

2. MAINT DDR
SYS CONS
IN 123 DASD
OUT D3F TAPE
DUMP ALL
YES

3. After writing the first tape full, DDR want's the second one.
How to mount it? The rdev d3f is attached to MAINT, so can't mount from another 
machine/user. MAINT is "in DDR session" so can't mount from there either. 

4. The tapesystem is VTS using 3584 library.
Any ideas how to get the following output tapes mounted?
(I tested DDR output parm altape to write two tapes but then third tape was 
required and there I was again ... )

Thanks
Janne J

    

  


Think green - keep it on the screen.
This e-mail and any attachment is for authorised use by the intended 
recipient(s) only. It may contain proprietary material, confidential 
information and/or be subject to legal privilege. It should not be copied, 
disclosed to, retained or used by, any other party. If you are not an intended 
recipient then please promptly delete this e-mail and any attachment and all 
copies and inform the sender. Thank you. 

Reply via email to