At 9:46 AM -0500 2005-09-22, Brian Elliott Finley wrote:
> I'm very interested in the solution you chose, and in how well it works
> for you.
Sorry, I missed the original message.
> I'm in the process of designing a fault-tolerant mailman installation,
> and any wisdom from experience would be very useful.
Generally speaking, I don't think that many people try to do
this. You can off-load the web processing with multiple front-end
web proxies, and you can off-load the mail processing with multiple
inbound and outbound mail servers (usually best kept as separate
groups, since the anti-spam, anti-virus, and certain other
requirements are different for inbound versus outbound use), and the
rest is just Mailman itself.
But, if you do want to go the whole nine yards, the Mailman
programmers have been pretty careful to keep everything relatively
NFS-safe, so you should just be able to have an NFS filesystem which
is then mounted on all of the Mailman servers (preferably served by a
high availability/redundant NFS server cluster). Doing things over
NFS will slow down individual operations, but the overall aggregate
throughput might be able to be higher, if the NFS server has the
right architecture and you spread the load out across enough Mailman
servers, etc....
>> Any thoughts on what is the best way to synchronize mailman databases
>> between to linux boxes?
Mailman doesn't really use databases. At least, not
out-of-the-box. There are third-party unsupported patches to allow
database member adapters, but that's about it.
Mailman stores pretty much everything into Python "pickle" files,
which is a particular binary Python-specific file format that can be
quickly saved and re-loaded in the native binary format.
>> If my alias_maps are defined the same in /etc/postfix/main.cf are the
>> same on both hosts can I just duplicate my /usr/lib/mailman and
>> /var/lib/mailman directories?
Duplicate? What do you mean by "duplicate"? Just once? If so,
then the directories would get out-of-sync. You'd have to keep all
the directories in sync somehow.
In a nearline backup method, something as simple as rsync would
suffice, and would potentially only lose any information that was
written since the last rsync.
In a production OLTP type of environment, you'd have to use
shared filesystems between the sets of machines, and you'd have to
make sure that those shared filesystems implement all the necessary
cluster-wide locking facilities, etc... to keep Mailman working
correctly.
--
Brad Knowles, <[EMAIL PROTECTED]>
"Those who would give up essential Liberty, to purchase a little
temporary Safety, deserve neither Liberty nor Safety."
-- Benjamin Franklin (1706-1790), reply of the Pennsylvania
Assembly to the Governor, November 11, 1755
SAGE member since 1995. See <http://www.sage.org/> for more info.
------------------------------------------------------
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe:
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org
Security Policy:
http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq01.027.htp
