Forwarded message: > >As far as human delays go, we have no difference of opinion. A couple of >the mailing lists that I manage used to have significant delivery delays >for some users, apparently caused by the behavior you describe above. I >fixed the problem by dumping Sendmail in favor of Postfix. Its queue >manager is quite efficient (see http://www.postfix.org/qmgr.8.html for >details). Postfix also opted to send each recipient a separate email and has the added plus of doing its queueing in RAM instead of on disk. Postfix even includes limits on how many emails to deliver per hour to a downed site which is just coming back up. Otherwise all the pending deliveries can flood the receiving machine and possibly bring it down again. Postfix is definitely the way to go for sites with several mailing lists to serve. > >BTW, I hadn't heard of fastmail before. Is it similar to bulk_mailer >(ftp://cs.utk.edu/pub/moore/bulk_mailer/)? > I have heard the name 'bulk_mailer' but had never used it. When Drew made the request I started looking for it on RedHat 7.0 but found 'fastmail' instead. From what I have read about 'bulk_mailer' they do seem to behave the same way. The arguments given may differ though.