[Mailman-Users] mailman clustering?

[Pat Riehecky]

I am trying to build a nice failover system for random parts of our
infrastructure.  Does anyone know how to force mailman to replicate
changes over to another identically configured system?

I thought about using rsync for this but I am trying to keep the sync
triggered by updates on the primary server rather than just seeing if in
the last few minutes any files have changed.  Ideally I would like to
round robin the primary and the secondary to split the load up, but I
cannot do that without being sure that they are kept in sync with each
other.

How close to possible is this idea?  Is there something I should read?
I checked the FAQ and didn't see anything...

Pat