[prev in list] [next in list] [prev in thread] [next in thread] 

List:       info-cyrus
Subject:    Re: [POLL] virtual domains and Murder
From:       Fabian Fagerholm <fabbe () paniq ! net>
Date:       2002-07-27 20:24:23
[Download RAW message or body]

On Fri, 2002-07-26 at 10:49, Michael Fair wrote:
> First, this thread is explicitly talking about 
> IP based virtual domains in relationship to a 
> Murder.
[...]
> I fail to see how your comments relate to that 
> topic at all.

Read more carefully: it's not explicitly about IP based virtual hosting
and Murder, it's about virtual hosting and Murder in general.
Also, Ken Murchison explicitly asked people to tell what kind of ideas
they had with regard to Murder. So I think my post is right on topic.

> > My company has been testing an idea that is a little bit more wild:
> > load-balancing by network topology using Murder. 
> 
> I fail to see why this requires a murder versus
> just an IMAP proxy like perdition. 

I did not say it requires a Murder. I said it could be done using a
Murder. I think there are certain advantages of doing it this way,
although I'm willing to admit it might be one of the harder ways of
doing it at this time. There are several other ways that are much easier
at the moment, but in my opinion many of them are not as scalable or
elegant, or do not offer the potential for easy administration as with
Cyrus Murder with virtual domain support (and perhaps a number of other
features yet to be invented).

> Or you can have one murder at the home office with a large
> pipe and all satellite offices have a local IMAP proxy
> which will act just like a web proxy in that the mailboxes
> will be cached locally at the proxy and sync with the
> murder at the home office.  This is better in that the
> end user can change offices and the system setup doesn't
> need to change at all.

One of the open issues with this idea has been how a cache might affect
performance. Because this thread was not about caching, I refrained from
bringing up the subject - but I would certainly like it if Murder
front-ends could cache.

Just two quick comments:
1. I don't think it's sensible to start thinking about caching in Murder
right now.
2. IMAP clients already implement a lot of caching. Quite a little
information actually changes in mailboxes, with the exception of new
messages popping up in the inbox (or other boxes if you use sieve). They
need to be downloaded in any case, and I can't see why that couldn't
happen the usual way when the client requests the messages. An
intermediate cache won't affect this.

> Again why would they want to host the backend when
> a simple proxy would do... This is especially important
> for mobile users who will roam around a lot.  Having
> their mailbox's home jump from server to server while
> they roam around seems to me to be asking for trouble.

You probably misunderstood me. Mailboxes wouldn't "jump from server to
server". They would mostly stay put at the most suitable location.

> I fail to see how an SRV record is going to be any
> different than a standard A record to determine the
[...]

Yes, A or CNAME or anything else you can think of could do just as well.

> > Of course, there are a number of issues such as DNS SRV support in
> > clients, easy and safe moving of mailboxes between backends and so on.
> 
> Both of which are solved by using proxies instead of
> real backends.  If SRV records do indeed prove to be
> useful for this, then you only need ensure that the
> proxies support SRV records.  Since the choice in proxy 
> software is something you have control over, and in the 
> Open Source world even source control over, it is both
> realistic and doable to use SRV records for the task
> of proxies determining what server to contact...

If I were a nitpick, I could say that a proxy doesn't solve the issue of
lack of DNS SRV support in mail _clients_. :)
But you obviously don't see the benefits and potential of Murder in the
same way as I do, and you have a different solution to the problem which
is very fine. I was just interested in discussing Murder.

Cheers,
Fabian Fagerholm


[prev in list] [next in list] [prev in thread] [next in thread]