former03 | Baltasar Cevc wrote:
Hi everybody,
thanx to everybody who contributed for this great piece of server
software (and the nice support on the mailing list); explicitly thanx to
Ken for everything and David for the replication code!
Two questions, as I'm not able to jump into the code deeply enough to
answer them myself...
1. Is it possible to have a server be master and client at the same time
- with the effect of being able to use 2 servers at the same time for
IMAP connections?
Not (easily) with the current code, unless you have a way external to
Cyrus to keep track of which server is the master for a given mailbox.
You can NOT serve mailboxes/messages from a machine which is replicated.
I plan to add code, probably using MUPDATE, which will allow load
balancing across multiple replicated servers.
2. What gets replicated. Is it just messages? Or also the sieve things?
I've seen some sieve routines in the replication code, however, as it's
all in the imap directory, I am not sure...
Anything related to IMAP and Sieve (mailboxes, mailbox annotations,
messages, message flags, Sieve scripts, etc). Any non-critical data
such as deliver.db or tls_sessions.db is not replicated.
--
Kenneth Murchison
Project Cyrus Developer/Maintainer
Carnegie Mellon University
----
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
