[Info-vax] email forwarding

[Phillip Helbig---remove CLOTHES to reply]

In article <gpih5p$r8b$1 at online.de>, helbig at astro.multiCLOTHESvax.de
(Phillip Helbig---remove CLOTHES to reply) writes: 

> In article
> <8cb8c165-18eb-42b0-82eb-bcd477c64e14 at z1g2000yqn.googlegroups.com>,
> sapienzaf <sapienza at noesys.com> writes: 
> 
> > On Mar 14, 3:03 pm, hel... at astro.multiCLOTHESvax.de (Phillip Helbig---
> > remove CLOTHES to reply) wrote:
> > > From the account whose mail should be forwarded:
> > >
> > >    MAIL> SET FORWARD <address to be forwarded to>
> 
> Useful for an individual user who wants to forward his mail elsewhere.
> 
> > > From a priviledged account:
> > >
> > >    MAIL> SET FORWARD/USER=foobar <address to be forwarded to>
> > >
> > 
> > No additions from me.  That is the method I use.
> 
> Which is fine for normal cases, especially when setting up aliases.  
> 
> $  DEFINE/SYSTEM FOO BAR
> 
> The logical name can be useful if the forwarding address is changed
> often. With MAIL> SET FORWARD, one has to go into the MAIL utility. 
> From a procedure, this means writing a temporary procedure (I/O) and
> executing it (more I/O).  With a logical name, it is all done in memory.
> IIRC, it has to be in the SYSTEM table.  It would be nice if it could be 
> in LNM$SYSCLUSTER.  Presumably, MAIL looks explicitly in the SYSTEM 
> table for security reasons, and LNM$SYSCLUSTER came much later.
> 
> The distribution list in TCPIP$SMTP_COMMON (which I've defined as a 
> search-list logical name; the second translation is the default one 
> (SYS$SPECIFIC:[TCPIP$SMTP]) and the first one is on a non-system disk) 
> is, as far as I know, the only way to forward email to a list.  It is 
> also, as far as I know, the only way to forward email while keeping all 
> the headers in one block, so that they will be recognised as headers.

Something else about this method: if it contains an internet-style 
address on the same machine, the email will NOT be delivered.  However, 
email to local addresses (e.g. VMS usernames) will be.  Bug or feature?