Conference azur::mcc

    
  >>>  3.  #2 worked from the FCL.
    
    
    If it works for FCL not for Iconoic map, then it is  NOT a MCC_DNS
    problem.  May be there is some restriction in the ICONIC Map.
    
    
    Jesuraj
    
    
    

    
    I'm sure that the behind the scenes of this isn't a revelation to most
    readers of this conference but -
     
    MCC_COMMON:MCC_DNS_SETUP.COM runs SYS$SYSTEM:MCC_DNS_SETUP.EXE 
    and passes it the parameters -
    
    	p2 - file name
    	p3 - namespace:.DNA_Node
    	p4 - IDP value
    
    SYS$SYSTEM:MCC_DNS_SETUP.EXE then creates the phase4 node registration
    command files using SYS$SYSTEM:NETNODE_REMOTE.DAT but it creates the
    registration commands in this format -
    
$ !
$ ! Format for a registration command:
$ !
$ !   REGISTER NODE4 node-full-name SYNONYM=phase-iv-name
$ !
$ ON WARNING THEN EXIT
$ MANAGE/ENTERPRISE
REGISTER NODE4 comms_ns:.dna_node.HIDRT1 SYNONYM=HIDRT1 !IDP=49:: ADDRESS=63.0300 
REGISTER NODE4 comms_ns:.dna_node.NITWIZ SYNONYM=NITWIZ !IDP=49:: ADDRESS=63.0366 
REGISTER NODE4 comms_ns:.dna_node.COMMS  SYNONYM=COMMS  !IDP=49:: ADDRESS=63.0367 
REGISTER NODE4 comms_ns:.dna_node.NITDEV SYNONYM=NITDEV !IDP=49:: ADDRESS=63.0369 
REGISTER NODE4 comms_ns:.dna_node.DECDEV SYNONYM=DECDEV !IDP=49:: ADDRESS=63.0371 
REGISTER NODE4 comms_ns:.dna_node.NITMG2 SYNONYM=NITMG2 !IDP=49:: ADDRESS=63.0373 
REGISTER NODE4 comms_ns:.dna_node.MYJAVA SYNONYM=MYJAVA !IDP=49:: ADDRESS=63.0872 
    
    
    This forces a user, from the IMPM, to add Phase4 entities to a domain
    in this format-
    
    	dna_node.COMMS
    
    "...entered exactly as it was registered".
    
    If the user doesn't enter them this way, the Historian and PA don't
    work correctly and the user doesn't have a clue why.
    
    If you use  the FCL to add node4 entities to a domain the above
    restriction does not seem to hold.  You can go to the IMPM and the
    historian  seems to work fine.
    
    I agree that the problem would seem to be in the IMPM but,    
    SYS$SYSTEM:MCC_DNS_SETUP.EXE is breaking the rule of mixing case
    for a Phase4name.  I modified MCC_COMMON:MCC_DNS_SETUP.COM so it
    passes P3 to SYS$SYSTEM:MCC_DNS_SETUP.EXE in all uppercase but, it  
    still writes it out in lower case.

    The Phase4 use doc on page 2-21 lists a node reg command file in all
    uppercase but, that's not the way it is.
    
    My only regret is that I didn't see this before SSB.
    Any comments on recommending to customers that the node registration 
    command files be edited prior to executing them and changing dna_node 
    to DNA_NODE?  That is if a namespace isn't already in place and the 
    files have to be edited to reflect the directory structure for any 
    existing Phase4 nodes.
    
    /bill
    

I am a bit confused.  Why is it a Iconic Map problem?  The Iconic Map passes
EXACTLY what the user enters across the call interface.   It does not make
any assumptions based on the case of the data.

It is the managagement modules and some MCC routines that are having the
difficulties with respect to case.  If the user stores it in lower case, 
then accesses it in upper case, the user is told by the management module 
(via the presentation module) that the data could not be found.  If the 
management module does not care about the case, it should convert it to 
either all upper or lower case before storing the data.  The same
should be done prior to retrieving it.  

A generic presentation module should not make assumptions about case
sensitivity.  It should always pass it along as entered.  The management
modules know what should be done with it.  After all the management module
is the one that wants the user input.

/Jim

    Jim, it is a DECmcc V1.1 problem. Users of the Iconic Map see it.
    Users have one product, not a bunch of independant products. Users
    don't know about nor care about call interfaces and which module does
    what about what.
    
    It is simply a case of part of the product being out of compliance with
    a restriction that is documented in the release notes.
    
    We, the users or testers of the product arn't trying to point fingers
    at one or another management module (the IMPM is an MM too). We are
    pointing out issues that need to be addressed.
    
    s/rob

    Oh, and for clarification, the problem is a case sensitivity in the MIR
    routines that the Historian and Exporter use. They need to be fixed for
    V1.2.
    
    s/rob

    I second the comment in .4.  I'm looking at it from the perspective 
    of a user and am not pointing fingers.  
    
    In the Network Management Using DECmcc course I will include material
    on editing the nodes command files and changing dna_node to DNA_NODE.
    
    bill