I have a Unix server and offer services to clients. On the server
clients offer to their customers legitamate newsletters that require
double-opt. I currently use sendmail and each user has their own IP,
but the server sends emails out on one IP. My concern is this. Even
though they are opt in newsletters and emails, what happens if someone
on the server is placed on a spam list (even though they are not
spamming). What can I do? Should we change the server's IP frequently
to prevent being placed on spam list? Is there a way to change
sendmail so each site uses its own IP instead of the server as a
whole? Any suggestions are welcomed. Once again, we do not send spam
and everything is optin with the ability for the customer to be
removed from email list with one click or shown with another click
when and where they signed up for the newsletter - zero spam.

If I recall correctly, most Blacklisters won't blacklist unless there
is an open relay. (Send mail through SMTP without authorization.)
Someone like Yahoogroups may have an identical issue to yours, but
they won't be necessarily permanently blacklisted because there is
adequate valid traffic.

Some blacklisters will be more general about whom they filter than
others, but if you check http://www.ordb.org/ or
http://www.ordb.org/lookup/?host=your.ip.address.here periodically,
you can get a feel for how you're doing.

You may not be able to prevent an individual from blocking all from
your IP, but you can work to make sure you don't get on noted spammers
lists. (and, yes, changing your IP address periodically can help...)

The best way I found to not be black listed is to use your ISP's mail
server as a relay.  Its really easy to have sendmail relay all email
through another server.  Most ISPs trust anyone on their network so
simply add the following line to sendmail config file.

define(`SMART_HOST',`smtp.your.provider')dnl