On Fri, Jul 30, 2004 at 05:16:53PM -0700, Mark Dadgar wrote: > On Jul 30, 2004, at 4:24 PM, Ed Wilts wrote: > >>I do not use sendmail [yet - will it work with > >>comcast?] > > > >Yup. You should make sure that sendmail.mc has a SMART_HOST entry > >pointing to Comcast's mail server though - this prevents other hosts > >from blocking your outbound mail. > > If you run a mailing list, even a small one, it's distinctly possible > that your ISP will get grumpy about the load you are putting on their > (apparently candy-*ssed) servers. > > Ask me how I know. I happen to have Comcast and so far they don't care. I push through a reasonable amount of mail and do have a very low volume 100-member mailing list. Whenever mailman sends a batch out, I see the throttle messages from Comcast's server. This introduces a delay but it's perfectly fine for what I'm doing - I'm not trying to run redhat-list or anything. What I've seen over the years here is that as long as you're reasonable, you stay under their radar. I have heard lots of horror stories from other ISPs who are extremely picky about their terms and conditions. Even the web and e-mail servers I run don't see to bother Comcast. I keep them low volume and for non-profit organizations. I scan all my e-mail, including those messages I simply act as a virtusertable entry for. I don't run IIS and don't open myself up for massive outbound attacks. So far, so good. I see Comcast e-mail scanning regularly looking for open relays. I like that in an ISP - it shows they care about the global problem. They know I'm running an SMTP server but don't do anything about it. I'm not in any hurry to switch ISPs. I still suspect my overall network traffic is considerably less than those people downloading CDs and MP3s all day long. -- Ed Wilts, RHCE Mounds View, MN, USA mailto:ewilts@xxxxxxxxxx Member #1, Red Hat Community Ambassador Program -- redhat-list mailing list unsubscribe mailto:redhat-list-request@xxxxxxxxxx?subject=unsubscribe https://www.redhat.com/mailman/listinfo/redhat-list
