lmtp versus deliver
Philip Chambers
P.A.Chambers at exeter.ac.uk
Mon Nov 8 07:07:56 EST 2004
On Fri, 05 Nov 2004 15:00:34 -0500 Ken Murchison <ken at oceana.com> wrote:
> Philip Chambers wrote:
>
> > This is cyrus-imap.2.2.8 with exim-4.41:
> >
> > I had been using lmtp to do normal deliveries but deliver to deliver to specific
> > folders (as for spam being diverted to a spam-folder.
> >
> > I noticed in my exim logs that at busy times I was getting a few failures from the
> > deliver program (logged as error code 75 or 65 by exim). Can someone tell me what
> > these two codes mean?
> >
> > On reflection I thought that I might be able to use lmtp to deliver to specific
> > folders instead by using "user.folder-name at dom.ain" as the address. I tried this
> > and it worked. An advantage of this is the benefit of multiple deliveries in one go.
> >
> > However, I have now found a difference which it would be nice to remove:
> >
> > Using deliver to deliver to "user.folder", if the folder does not exist, it delivers
> > to the user's inbox. Using lmpt, if the folder does not exist, it rejects the
> > message.
> >
> > Is it possible to get lmtp to deliver to the inbox if the specified folder does not
> > exist?
>
> Either I'm confused or you're using the incorrect syntax. To deliver to
> a user-specific folder via LMTP, the address should be
> user+folder at dom.ain. If the folder doesn't exist (or doesn't allow
> posting by the [un]authenticated user, then the message *will* go to the
> user's INBOX.
Thanks for that information, as you say, using plus as the separator delivers to the
folder if there is a suitable acl or to the inbox if not.
Using a dot to separate the username and foldername certainly works and the delivery
takes place without any acl on the folder.
Phil.
---------------------------------------
Phil Chambers (postmaster at exeter.ac.uk)
University of Exeter
---
Cyrus Home Page: http://asg.web.cmu.edu/cyrus
Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu
List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
More information about the Info-cyrus
mailing list