| sean finney on 22 Jan 2005 21:55:33 -0000 |
|
On Fri, Jan 21, 2005 at 10:47:24PM -0500, Michael C. Toren wrote: > On Fri, Jan 21, 2005 at 08:33:40PM -0500, sean finney wrote: > > it's an abuse of the smtp protocol, for starters. > > I disagree; it's completely within the protocol, and a clever hack to > distinguish between fire-and-forget mailers and true store-and-forward > mail servers. it's definitely a clever hack, but i'd still argue that it's an abuse of the protocol--you don't have to break a protocol to abuse it... > While it does increase the resources required for a remote host to > send me mail, it's hard to make the argument that it doubles load. > The network overhead for example is tiny, as 4xxx temporarily failure > messages are returned after the RCPT TO, well before the DATA segment. > Additionally, only the first message for the <IP,sender,recipient> > triplet is affected -- afterwards, the triplet is whitelisted. okay, you called me out on an exaggeration, it doesn't double the total overhead. still, it's more than you suggest. for example, if my mail server is sending mail to a server that implements greylisting, i have to store the message in a temporary queue until my mail server attempts to retry. so, instead of the queue liftetime of a message being a few seconds, it can become minutes or hours depending on the mta. if you have a busy mail server for a large user-base, this is a huge impact on resources. sean Attachment:
signature.asc ___________________________________________________________________________ Philadelphia Linux Users Group -- http://www.phillylinux.org Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce General Discussion -- http://lists.phillylinux.org/mailman/listinfo/plug
|
|