Thanks Elardus.  I will check around with our MVS gurus to see if I can remove 
the command.
--------------------------------------------
On Thu, 8/12/16, Elardus Engelbrecht <[email protected]> wrote:

 Subject: Re: EXPLANATION ABOUT VALIDATE WHEN DOING A DS QD
 To: [email protected]
 Received: Thursday, 8 December, 2016, 11:23 AM
 
 John Dawes wrote:
 
 >Thanks Elardus.  
 
 You're most welcome! ;-)
 
 First thing first - perhaps you should remove that command
 from the batch job you discovered.
 
 
 >What does "Host view  is consistent with subsystem
 view of control unit and device  function status.' I am
 unable to understand IBM speak.
 
 Neither me...
 
 
 >Could you elaborate or give an example as to what it
 means?
 
 I wish I could elaborate on that, but as a previous storage
 admin + MVS admin, I remember that during the configuration
 of the devices and later VARY ONLINE, somehow the definition
 of the control units and their devices are placed somewhere
 in a table after the system and control units have a 'chat'
 about defined devices / control units. 
 
 Other things considered: how the device is shared, what
 paths and what control units are connected to what LPAR /
 SysPlex, how the caching is defined/working.
 
 What I know, is when you IPL, the operating system use the
 IODEVICE macros for example and scan the IO config or using
 your setup in HCD. Only devices which are initialized and
 varied online are available for use. The system is 'talking'
 with each control unit and then establish tables for each
 devices and control units. Of course your CE should get
 those devices ready for use.
 
 But see Bill Woodger's excellent comments about system
 control blocks.
 
 So if a device is 'boxed' or in a similar status, the
 operating system, the hardware on the computer and control
 unit are not 100% agreeing with the real status of a device.
 Hence those DS commands as an aid amongst other things for
 you.
 
 One example I know is, if you vary a device offline on one
 LPAR, but the device is in use in the same LPAR or other
 LPAR in same or other SysPlex. Then you get some
 inconsistency. This is messy enough that I locked up the
 VARY command in RACF.
 
 Above is based on my frail and faint memory as storage
 admin. 
 
 Perhaps others can chime in with better examples or pointers
 to good documentation.
 
 Groete / Greetings
 Elardus Engelbrecht
 
 ----------------------------------------------------------------------
 For IBM-MAIN subscribe / signoff / archive access
 instructions,
 send email to [email protected]
 with the message: INFO IBM-MAIN

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to