User unknown in virtual alias table

Drew Marshall drew at THEMARSHALLS.CO.UK
Sun Mar 13 11:42:37 GMT 2005

    [ The following text is in the "windows-1252" character set. ]
    [ Your display is set for the "US-ASCII" character set.  ]
    [ Some characters may be displayed incorrectly. ]

Steen, Glenn wrote:

 -----Original Message-----
From: MailScanner mailing list on behalf of Drew Marshall
Sent: Sat 3/12/2005 11:05 PM
Subject:      Re: User unknown in virtual alias table
Steen, Glenn wrote:


 Hair of dog, eh Drew.... Is that good against hangovers ....? (I'll be

 needing *something* tomorrow, or else my choirleader will.... well, let's
not go there.... Oh, not to mention the rotations JSBach will be doing in
his tomb...:-).

 Now I'm not sure if you have picked up on the English colloquialism
'hair of dog' not it's literal phrase. Hair of dog is referred to as the
first beer in the bar the following morning (For example) as opposed to
the fur of a canine :-) . One will add to your hangover but if taken in
sufficient quantity will help you forget it (Although pickle your
liver!) the other will make you feel ruff (Sorry couldn't resist :-) ).

 Ah. No that particular turn of phrase had eluded me.... I'll be sure to
remember it:-)
Wouldn't be doing me much good anyway.... need to be sharp to sing


 ... Would it be entirely unacceptable to "fix" this in documentation...

 "don't do that" type of thing?

 I don't know how tricky it will be to fix the code TBH (I code like a
sysadmin ;-) ). There is no reason why the docs can't read that for
Postfix you can't enter virtual aliases. I'm not sure I personally like
it as I prefer things to work (TM) but it is a work round. It's a pain
that Postfix insists on doing it's virtual alias lookups in the cleanup
stage after either pickup (Injected messages) or smtpd (SMTP input)
before MailScanner gets to look at it. This document here details it rather
well. See the table about 1/3rd down the page. That specifies which
Postfix component does the address rewriting for aliases.

 Yes, I know. But "doing it right" would mean not to just fiddle with the queue
files and use "normal methods" for injecting it back in, or playing cleanup in
MW (which seems horrendous). Or perhaps I'm missing something in my somewhat
hubgover state?-).

Well as I see it there are 2 options, one is to change the docs, the
other is to change the hold file regex to exclude direct injected mail
(From localhost) and get MS to queue inject the forwarded file so it is
not re-scanned. Neither is particularly pretty or an ideal solution. My
preference would be your idea to change the docs as after all if the
system admin is setting the forward option in MS then (s)he should also
know the unaliased address of the recipient.

Maybe Julian has a better idea??


In line with our policy, this message has been scanned for
viruses and dangerous content by MailScanner, and is
believed to be clean. ------------------------ MailScanner list
To unsubscribe, email jiscmail at with the words:
'leave mailscanner' in the body of the email.
Before posting, read the MAQ (
and the archives (

Support MailScanner development - buy the book off the website!

More information about the MailScanner mailing list