murder and autocreate (I know it is not supported)

[Greg A. Woods]

At Mon, 15 Jun 2009 07:42:14 -0400, Dave McMurtrie <dave64 at andrew.cmu.edu> wrote:
Subject: Re: murder and autocreate (I know it is not supported)
> 
> Exactly.  The point I was trying to make is that we already have a need 
> for some system to be able to connect to our IMAP server for the purpose 
> of deleting mailboxes, so having that same system connect to our IMAP 
> server to create mailboxes seems to make perfect sense.

That's all fine and well for those who wish to automate deletion of
mailboxes.

Note that I'm not saying anything about preventing or removing the
ability to manually or programmatically create mailboxes -- just that
Cyrus in many (even most by numbers?) environments _must_ have the
ability to automatically create mailboxes on demand if it's to be easily
managed without having a vast majority of those installations also have
to craft or find some additional custom mailbox management tool that's
more likely to be an ugly hack than a secure and clean design.

Even if you do have your user management system create mailboxes You
still need to have your MTA validate addresses.  And, unless you do
mailbox creation before user creation, you could still end up with a
window of vulnerability where the authentication database contains the
user account and the MTA presents a message to Cyrus before there's a
mailbox ready to receive it.  Perhaps this will normally be a tiny
window but I have actually seen an ISP account management system
reliably generate the welcome message to a new user before it finished
creating the new mailbox and thus the welcome message always bounced.
Yes it was a stupid design, but that's what happens when the underlying
systems are not first designed cleanly and elegantly.

Not having a built-in basic way for Cyrus to automatically create
mailboxes for valid users is also a poor design.  Several good
well-tested patches to enable this feature had been available for a very
long time (years?) before the "murder" feature was added in the first
place, and I have a firm conviction that if the autocreate feature had
been added to Cyrus when it was first made available then the design of
the clustering mechanisms would have supported it properly from the
beginning too.  I'm not even sure I understand the difficulty with it
now -- if the cluster front-end knows how to direct a user access to the
appropriate backend, then so can it direct an initial delivery for
mailbox creation.

-- 
						Greg A. Woods

+1 416 218-0098                VE3TCP          RoboHack <woods at robohack.ca>
Planix, Inc. <woods at planix.com>      Secrets of the Weird <woods at weird.com>