[comp.unix.questions] Sending mail with explicit INTERNET address

mercer@npdiss1.StPaul.NCR.COM (Dan Mercer) (08/16/90)

I have accounts on two different systems tcp/ip'd together across
an Ethernet backbone.  Neither node is on the backbone,  but the
networks they are on are gated to the backbone.  On my home system,
where I have sysadmin privileges,  the remote system is in my
/etc/hosts file.  On the remote system,  where I'm a guest,  my
home system has not been put in their /etc/hosts file despite
numerous requests.  I can ftp and telnet from the remote to my
home system by using explicit decimal dot addressing.  My problem is,
how do I mail stuff to myself by the same method?

TIA
-- 

Dan Mercer
Reply-To: mercer@npdiss1.StPaul.NCR.COM (Dan Mercer)
"MAN - the only one word oxymoron in the English Language"

katsu@sra.co.jp (WATANABE Katsuhiro) (08/20/90)

In article <104@npdiss1.StPaul.NCR.COM> mercer@npdiss1.StPaul.NCR.COM (Dan Mercer) writes:

> I have accounts on two different systems tcp/ip'd together across
> an Ethernet backbone. Neither node is on the backbone,  but the
> networks they are on are gated to the backbone.  On my home system,
> where I have sysadmin privileges,  the remote system is in my
> /etc/hosts file.  On the remote system,  where I'm a guest,  my
> home system has not been put in their /etc/hosts file despite
> numerous requests.  I can ftp and telnet from the remote to my
> home system by using explicit decimal dot addressing.  My problem is,

> how do I mail stuff to myself by the same method?

 Though the subject is "Sending mail with explicit INTERNET address",
I think, there are more better ways. I cannot see why you wish mails to
be delivered directly(without conventional mail routing). It may be
useful to contact with administrators of your system.

  ..... anyway, RFC822 says that

RFC>      addr-spec   =  local-part "@" domain        ; global address
RFC>      domain      =  sub-domain *("." sub-domain)
RFC>      sub-domain  =  domain-ref / domain-literal
RFC>      domain-ref  =  atom                         ; symbolic reference
RFC>      domain-literal =  "[" *(dtext / quoted-pair) "]"

RFC>             o   Square brackets ("[" and "]") are used to indicate the
RFC>                 presence  of  a  domain-literal, which the appropriate
RFC>                 name-domain  is  to  use  directly,  bypassing  normal
RFC>                 name-resolution mechanisms.

RFC>         A domain-ref must be THE official name of a registry, network,
RFC>         or  host.   It  is  a  symbolic  reference, within a name sub-
RFC>         domain.  At times, it is necessary to bypass standard  mechan-
RFC>         isms  for  resolving  such  references,  using  more primitive
RFC>         information, such as a network host address  rather  than  its
RFC>         associated host name.
RFC> 
RFC>         To permit such references, this standard provides the  domain-
RFC>         literal  construct.   Its contents must conform with the needs
RFC>         of the sub-domain in which it is interpreted.
RFC> 
RFC>         Domain-literals which refer to domains within the ARPA  Inter-
RFC>         net  specify  32-bit  Internet addresses, in four 8-bit fields
RFC>         noted in decimal, as described in Request for  Comments  #820,
RFC>         "Assigned Numbers."  For example:
RFC> 
RFC>                                  [10.0.3.19]
RFC> 
RFC>         Note:  THE USE OF DOMAIN-LITERALS IS STRONGLY DISCOURAGED.  It
RFC>                is  permitted  only  as  a means of bypassing temporary
RFC>                system limitations, such as name tables which  are  not
RFC>                complete.

  Now, assume the IP address of your remote host to be "1.2.3.4",
and login name to be "guest". Then, try
        mail "guest@[1.2.3.4]"

  But, there is no guarantee that your system(to say, /usr/lib/sendmail.cf)
can handle notations like [x.x.x.x] correctly. (You can check with
sendmail -bt.)

# Sorry for my poor English.

----____----____
WATANABE Katsuhiro      Software Research Associates, Inc. Japan.
Not execute, but evaluate.