On Monday 31 October 2005 10:09, Dan Mahoney, System Admin wrote: > On Sun, 30 Oct 2005, Larry Smith wrote: > > ((cut to bottom to keep in line)) > > > On Sunday 30 October 2005 15:18, Dan Mahoney, System Admin wrote: > >> On Sun, 30 Oct 2005, Phil Beynon wrote: > >> > >> I used to have "f*** off and die, spammer" in mine -- now I just take > >> what would match those blacklists and use them to feed my databases. > >> > >> Let me counter -- every one of the old netiquette guidelines (around > >> since the 1970's, email is now 25 years old) say not to SEND spam. So > >> if someone's going to violate those rules, well... > >> > >> -Dan > >> > >>> Is there any particular netiquette guidlines as to what should be put > >>> in sendmail access reject texts? > >>> > >>> Having been the receipient of a vast amount of really not terribly nice > >>> spam recently I've started really getting quite rude in the reject > >>> texts they get back - well quite offensive really! :-) > >>> > >>> Regards, > >>> > >>> Phil Beynon > >>> Sales director > > > > While I really do understand the thread, one thing to remember is that > > none of us is "perfect" - meaning what do you do when you get a "false" > > and tell either one of your customers, upstreams, or an "aol", "hotmail" > > or such to "f" off and die..... > > > > We need to remember that like our community that we "live" in > > (physically) we make the internet in some small way each day (or more > > particularly someone else's opinion of it) and as such should "temper" > > what we say in our error messages. > > > > I find that the more "useful" the error message, the more probability of > > getting something done (sort of the honey catches more files than vinegar > > concept). Does that mean the hardcore spammers are going to change their > > ways - probably not, but since most of them are using either hacked, > > hijacked or other connections and "false", forged or other addresses, > > that those "error" messages, bounces and such that end up going to a > > "real" person somewhere will at least have some "clue" as to what > > happened and not be immediately 'po'd' at you or me for the "rude" > > message that gave them no clue as to what or why (and in fact will > > probably view you/me as the "spammer" and request us be blocked by their > > ISP).... > > > You misunderstand. That response was reserved not for blacklists, but for > thousands of email addresses at a domain that only had ONE active user > with one published address. I was not telling hotmail or anyone else to > do anything -- I was talking to spammers. Admittedly, someone who > misspelled the single user name (which was also the user's first name) > would get that nasty message -- but there was NO reason for those messages > to be coming through. None. And I wanted something easy to spot in the > logs. That was it. > > The bandwidth used by throwing a 550 at the SMTP level was far cheaper > bandwidth wise than actually accepting the emails. > > For the record, those mails pushed my system load way above an acceptable > average due to the increased load via spamassassin. > > For the record, you asked about access reject tests, not DNSBL reject > text. > > -Dan
Hmmm, nope, I did not mis-understand, I just answered from a different perspective than was probably expected. I agree with you whole-heartedly - and do the same myself. A 5XX response and drop the connection(s) costs me a lot less than even worrying about any handling of the possible mail involved. None-the-less - the conversation was about the "content" of the reject message, whether it be a 5xx or bounce or DNSBL or other. And I still maintain my response that "courtesy" goes a long way and that there is seldom a really good reason to be "rude" or such - even in reject messages. Put another way, the person who has never, ever made a mistake (of any sort) can pretty much do what they want ...... The rest of us have to moderate our behavior on the off chance that the 212.x.y.z IP we really meant to add to the access file actually got entered as 12.x.y.z and we end up telling AT&T customers/servers/whatever to (paraphrased) "take a hike".... Was not intended to criticize you or your responses in any way - mearly to give all of us a perspective to "think" a bit about what language and such we put out there in our own spaces (whether error messages, web pages, or whatever)... -- Larry Smith SysAd ECSIS.NET [EMAIL PROTECTED] _______________________________________________ Cobaltfacts site list [email protected] http://list.cobaltfacts.com/mailman/listinfo.cgi/cobaltfacts
