[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference azur::mcc

Title:DECmcc user notes file. Does not replace IPMT.
Notice:Use IPMT for problems. Newsletter location in note 6187
Moderator:TAEC::BEROUD
Created:Mon Aug 21 1989
Last Modified:Wed Jun 04 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:6497
Total number of notes:27359

854.0. "Recycling of Default IDP 49::?" by LTLKNG::TAZBOY::ZIGLER (Tom Zigler DTN:432-7541) Mon Apr 01 1991 12:15

    I am receiving numerous customer questions regarding the following
    DECmcc IDP issue:
    
    "If I choose the default IDP (49::) today, will I still be able to
    RECYCLE this IDP to a different one, assigned to me by an OSI allocation
    authority, at some indeterminate future time?"
    
    What is our official position on this question and how much pain is
    involved if the answer is "yes"?
    
    Please advise.
    
    		\Thanks in Advance
T.RTitleUserPersonal
Name
DateLines
854.1YesTOOK::KOHLSRuth KohlsMon Apr 01 1991 12:5731
    
>    "If I choose the default IDP (49::) today, will I still be able to
>    RECYCLE this IDP to a different one, assigned to me by an OSI allocation
>    authority, at some indeterminate future time?"
>    
>    What is our official position on this question and how much pain is
>    involved if the answer is "yes"?
>    

	49:: is the official NO IDP IDP value.  (perhaps it should have
	been 42...(;-|))  The installation process asked for an IDP, 
	validates its form, and creates a logical name on VMS or environmental
	variable on ultrix (I'm guessing there) with the validated IDP. 
	
	I suspect that if someone was sure of the correct form of their
	newly assigned IDP, they could edit (on VMS) 

		sys$startup:MCC_LOGICAL_DIR.com

	and change the value of the logical name MCC_NODE_IDP.  They should
	also redefine the value of the MCC_NODE_IDP logical in the current 
	system logical name table.  I expect some similar procedure is being 
	designed for mcc on ultrix. 

	The derivation of the IDP is covered in the Installation
	Guides for both BMS and DIR kits.

	Unless there is more "pain" involving the Phase 5 AM, that ought
	to do it.

Ruth 
854.2problems with 49::NAC::ENGLANDMon Apr 01 1991 13:508
    It may not be a good default, because if your customers want to use
    OSI for connecting up multiple networks, the 49:: AFI doesn't support
    this.  You might want to investigate GOSIP-compliant AFIs, etc.
    There may not be a good default value at this time.  I'm very ignorant
    of OSI addressing, but you might want to talk to Stan Goldfarb.
    
    ben
    
854.349, the new answer to Life, the Universe, and EverythingFAFNIR::GOLDFARBStan Goldfarb, LKG1-3/L06Mon Apr 01 1991 18:1523
I was told the my name was being taken in vain (or perhaps that should
be "vein", since enough blood has been drawn on this subject).

The value "49" is the default because:

  1) Most small customers will not have to worry about interconnecting networks
     for a LONG time (if ever), and should be able to choose a default without
     thinking about it.

  2) There is no good default to use.  49 is as good as any other.

As for fixing it later to allow the network to be interconnected, I am currently
adding a function to NET$DNS_REGISTER (decnet_dns_register on ULTRIX) to help
with this.  NET$DNS_REGISTER is the node name registration tool.  The new
function will allow the network manager to go into the namespace and change
the IDP for all registered nodes.  Each node manager must, unfortunately, 
change the IDP on their own nodes, but at least there the work will be spread
out among lots of different people.  This will fix things up for both
Phase V and Phase IV nodes registered in the namespace.

I think Ruth already described how to make DECmcc happy in her ".1" reply.

/Stan
854.4Backtranslation directories ?CCIIS1::ROGGEBANDThu Apr 04 1991 06:0411
    Re: .1
    
    The DNA Backtranslation subdirectories all have names of the format 
    
    %X49.%X00<area # in hex>
    
    I expect you would have to copy the subdirectories to the ones under
    you new IDP, in order to let your Phase IV & Phase V nodes talk to each
    other ? 
    
    Philippe.
854.5moving backtranslation entriesCLAUDI::PETERSMon Apr 08 1991 16:5310
No you can not copy them, you will either have to use the tool Stan
mentioned or deregister and register everything.  The com files that 
your created to initially register everything can be used to register
the nodes with the new IDP.  Because DECmcc touches each node that 
it registers, moving from one IDP to another can be a long process.
So consider the initial IDP with care.  I suspect that the tool will
do what you want copy one DNA_BACKTRANSLATION tree to another and 
will not require the slow registration process, though it might have
to touch both the backtranslation link and the object to fix its address
towers. /Claudia
854.6CRONIC::LEMONSAnd we thank you for your support.Wed Feb 26 1992 15:183
What IDP value should be used by me, an internal Digital site?

tl
854.7Question5ONFLR::BONAZZOLIWed Mar 04 1992 12:535
      I have the same question as -1.  I have an internal node that I
    need to install MCC on asap, but I don't know what the IDP should
    be (if anything).  Can anybody help?
    
    Rich
854.8We use 49::TOOK::MINTZErik Mintz, DECmcc Development, dtn 226-5033Wed Mar 04 1992 15:575
The default (49::) is typically the correct answer for internal
systems, as far as I know.

-- Erik

854.95ONFLR::BONAZZOLIThu Mar 05 1992 12:232
      Thanks,
    Rich