Global identifiers for initiators

Bob Snively Bob.Snively at eng.sun.com
Mon Jan 2 22:20:28 PST 1995


Ralph,

In parallel with Charles' comments of a few seconds ago,
I believe that it is the responsibility of the "family
moving in" to discard or forward any incoming actions for which
it has no responsibility.  Similarly, the "family moving in" is
responsible for performing its own change of address and change
of account actions for all utilities.  If it fails to perform
any of the required actions, the family suffers both the consequences
and the culpability.  None of these actions is the responsiblity
of the postal system or the phone service provider across which
the communications take place.

I like your analogy.  It is very powerful.

Bob


>From @ncrwic.WichitaKS.NCR.COM:weber at star.enet.dec.com Fri Dec 30 02:22 PST 1994
>Date: Thu, 29 Dec 94 17:09:53 EST
>To: scsi at wichitaks.ncr.com
>Cc: weber at star.enet.dec.com
>Apparently-To: scsi at wichitaks.ncr.com
>Subject: RE: Global identifiers for initiators
>Content-Type: text
>Content-Length: 1872
>X-Lines: 33
>
>When considering the question of global identifiers for initiators, one 
>should think about the question, "What does an initiator global identifier 
>represent?"  Stated in postal-service terms, "Does the global identifier 
>represent the house or the people living there?"
>
>For a SCSI target device (disk, tape, etc.), the house and the people living 
>there are pretty much synonymous.  In the case of a non-removable medium, the 
>two are synonymous.  The contents of the media cannot change except as the 
>result of a transaction directed to the disk's global address (or a serogate 
>address).
>
>Initiators, on the other hand, can have one family move out and a whole new 
>family move in more or less on a whim.  It's done by shutting down operating 
>system one and booting operating system two.  A shutdown and reboot of the 
>same operating system has many of the same properties.  So, does the global 
>identifier represent the running operating system and all its running applica- 
>tions?  Or, does the global identifier represent the box that the operating 
>system is running in?
>
>My experience with VMS Clusters suggests that the latter type of representa- 
>tion is not very useful.  But, the former case is very difficult to define 
>and implement.  For example, the former case requires the operating system to 
>define and present its new global identifier (that is different from the last 
>boot) very early in the booting process.  Also, the operating system's global 
>identifier must be the same on all access ports, which might get very ugly 
>depending on which protocol layer defines the transmission methods for the 
>initiator's global identifier.
>
>I don't have a specific solution in mind here.  I just don't want the global 
>identifier debate to attach too much value to global identifiers.  I've been 
>there enough times to know that it doesn't work.
>
>Ralph::
>




More information about the T10 mailing list