RE: iSCSI Naming: WWUIs, URNs, and namespaces

["Douglas Otis" <dotis@sanlight.net>]

David,

Perhaps we are having a problem with semantics. A reverse DNS is perhaps not
the best name for this function if it has nothing to do with a reverse
lookup of an IP.  You are taking a domain and attempting to find related
services.

   Alias     Service
   -----------------------------------------------------------
   archie    archie [ARCHIE]
   finger    Finger [RFC-1288]
   ftp       File Transfer Protocol [RFC-959]
   gopher    Internet Gopher Protocol [RFC-1436]
   ldap      Lightweight Directory Access Protocol [RFC-1777]
   mail      SMTP mail [RFC-821]
   news      Usenet News via NNTP [RFC-977]
   ntp       Network Time Protocol [RFC-1305]
   ph        CCSO nameserver [PH]
   pop       Post Office Protocol [RFC-1939]
   rwhois    Referral WHOIS [RFC-1714]
   wais      Wide Area Information Server [RFC-1625]
   whois     NICNAME/WHOIS [RFC-954]
   www       World-Wide Web HTTP [RFC-1945]
   iscsi     iSCSI Server [RFC-xxxx] (Name Server via SCSI)

So from what I see from your example is iSCSI.Example.Com returns the WWUI
via SCSI that is then inserted into the domain to obtain the desired IP.  As
example [WWUI].example.com which is not World Wide Unique Identification but
at least unique to example.com.  Why are you describing this process as
reverse DNS?  If this is the process then I find the term reverse DNS
confusing.  You are referring to a conventional alias.  As this one machine
may not be able to carry the load for all clients or connect to all targets,
would you then break connection to the iSCSI machine and then make direct
connections to the LUNs?  Would a domain support multiple SCSI portals with
other names like iscsi1 or iscsi2?  Is that also to be a standard
convention?

Doug

> > No, I was not confused nor did I expect clients to turn a WWUI into an
> > address as obviously you already have an address to do the
> reverse DNS.  I
> > indicated the process as Name->IP->Reverse-Name and illustrated this
> process
> > by pointing to the technique that Microsoft used to obtain names from
> > machines.
>
> I'm afraid Doug is still confused.  For connection to a specific target
> identified
> using a Reverse DNS name, the sequence is:
>
> Reverse Name --> DNS name --> IP
>
> where the first association is handled by some sort of
> nameservice (such as
> the
> LDAP server that Doug favors) and/or static configuration and the
> second is
> the usual DNS lookup.  The first association has to be some sort of table,
> directory or database lookup, as extracting the DNS component of
> the reverse
> DNS name and trying to connect to it will not work in general.
>
> The only way to get from IP to the reverse name is to connect to the
> canonical iSCSI Target ("iscsi") and use the "Send Targets" command to
> get a list of targets visible to the initiator.  Spoofing concerns
> get handled at the stage where the Initiator connects to the Target
> (there's a full iSCSI login that may be riding on a secure IPSec or TLS
> connection, depending on how we decide to do this).  The Reverse
> Name is NOT being extracted from DNS.  This mechanism
> is analogous to the SCSI REPORT_LUNS command and the way
> that a Fibre Channel initiator downloads a set of targets from the FC
> nameserver as part of FC discovery -- both of these mechanisms
> work fine in appropriate environments, and neither has anything to
> do with NetBIOS.
>
> As for LDAP - writing a draft on how to use it would be a far more
> productive approach than complaining that the WG is ignoring LDAP.
>
> --David
>
> ---------------------------------------------------
> David L. Black, Senior Technologist
> EMC Corporation, 42 South St., Hopkinton, MA  01748
> +1 (508) 435-1000 x75140     FAX: +1 (508) 497-8500
> black_david@emc.com       Mobile: +1 (978) 394-7754
> ---------------------------------------------------
>
>