My 2 bits:
We use Senderbase Reptutation Scores and reject 80%+ of our connection attempts before accepting the message based on a negative score.
The minute that I can get all of my users consolidated into a single Active Directory Forest, I'll be using live LDAP to reject invalid recipients before I accept the message.
Soon after that we'll be implementing Directory Harvest Attack Prevention, that throttles connections after a theshhold of invalid recipients has been reached.
When we implemented RBLs several years ago, we took the "weighted" approach, where each list had an associated score. Using any particular single outside data source for complete control of the "pass / fail" was too risky for us.
Frank Hardwick
Tribune Enterprise Messaging
frank at tribune dot com
perfectlyphrank at yahoo dot com
Greg Mulholland <gmulholland@xxxxxxxxxxxx> wrote:
not sure i like the idea of live ldap lookups but updating the recipient table is a must..in my environments if you send an email to a non existing user i let the mta deliver the ndr..Greg----- Original Message -----From: John T (Lists)Sent: Thursday, October 19, 2006 8:19 AMSubject: [ExchangeList] Re: I NEED TO GRIPE!Those hosted solutions MUST NOW make changes to allow for either live LDAP lookups or an automated way of updating a list of acceptable addresses.John TeServices For You"Seek, and ye shall find!"-----Original Message-----
From: exchangelist-bounce@xxxxxxxxxxxxx [mailto:exchangelist-bounce@xxxxxxxxxxxxx] On Behalf Of Chris Wall
Sent: Wednesday, October 18, 2006 1:30 PM
To: exchangelist@xxxxxxxxxxxxx
Subject: [ExchangeList] Re: I NEED TO GRIPE!John ? I agree? In a perfect world that is the policy that we should all abide by?But what about those organizations that use Frontbridge or similar hosted SPAM/Content solutions and they are not able to verify if an email exists in a customer?s domain? The hosted center has to accept the email, filter it and then send to a recipient domain. If it is unable to send the message to the recipient domain b/c of recipient filtering or the address simply does not exist, then the hosted center sends out an NDR. This ?man in the middle?, which is growing in popularity will hurt many organizations because of SpamCop?s decision.Hopefully in the very near future, these hosted solutions will be able to perform live LDAP type lookups before accepting mail for delivery.
Regards,Chris Wall - MCSE + MessagingNAM Exchange AdministratorT (919) 460.3236F (919) 468.4889Global KnowledgeLEARNING. To Make a Difference.
From: exchangelist-bounce@xxxxxxxxxxxxx [mailto:exchangelist-bounce@xxxxxxxxxxxxx] On Behalf Of John T (Lists)
Sent: Wednesday, October 18, 2006 4:18 PM
To: exchangelist@xxxxxxxxxxxxx
Subject: [ExchangeList] Re: I NEED TO GRIPE!The biggest problem here that MANY over look is that you should not be accepting email for non-existent addresses!PeriodEnd of Story.I will back SpamCop on this issue. If you accept an email message and then are subsequently unable to deliver it to its final destination or to another server for delivery to its final destination then yes you are breaking RFC822. (Note that ?filtering? of the message in transit does not break RFC822 if such filtering is at the request of the receiver or his/her email provider etc etc blah blah blah)What has to be done is stop accepting incoming email for non-existent addresses. The connecting server is then given a 5.x.x return code and it is then up to that connecting server to notify the sender by NDR.John TeServices For You"Seek, and ye shall find!"-----Original Message-----
From: exchangelist-bounce@xxxxxxxxxxxxx [mailto:exchangelist-bounce@xxxxxxxxxxxxx] On Behalf Of Chris Wall
Sent: Wednesday, October 18, 2006 9:14 AM
To: exchangelist@xxxxxxxxxxxxx
Subject: [ExchangeList] I NEED TO GRIPE!Any one wanting to read or chime in, please feel free! I know all of you are e-mail admins, and you may have some thoughts on the subject as well.I am extremely disappointed with SpamCop.net ? one of the few blacklist sites that have ? rather, HAD a good reputation?Is any one else being affected by their actions of Blacklisting domains because they follow RFC822 and send NDR?s when a mail is not successfully delivered?Okay, here?s the overall story ? SpamCop sets up these ?HoneyPot? email addresses (whatever@xxxxxxx). SpamCop then sends e-mails out to many domains (illegitimate e-mail addresses ? basically acting as spammer?s themselves) and wait to see which domains send an NDR back to the ?HoneyPot? email address. If your domain follows RFC822 and sends the NDR, they blacklist the IP address of the server that sends the NDR.Their website (http://www.spamcop.net/fom-serve/cache/329.html#bounces) even details their stance on the issue. I have copied it below:?Q: Why not allow bounces? They are required by RFC822!
A: Originally, SpamCop made attempts to forgive misdirected bounce messages - to reject them as evidence of spam. However, there are two factors conspiring to force us to rescind this policy. First of course, is that these misdirected messages *are* spam as we define it (Unsolicited Bulk Mail). They are objectionable to recipients and can even cause denial of service to innocent third parties. Users of our blocking service want us to stop them.?I understand what they are trying to accomplish ? to prevent NDR?s from being sent to you when spammers ?spoof? your personal e-mail address. However, SpamCop is punishing domains that abide by all security standards for e-mail except for their ?rogue? approach to NDR delivery. Total BS in my opinion.Now of course, any domain could enable LDAP authentication on incoming e-mail and block NDR?s being sent when an e-mail address is sent to a non-existent e-mail address in your domain ? BUT, even excluding RFC822 rules requiring NDR?s on e-mails that are not successfully delivered, most organizations want to keep NDR?s enabled so that senders are aware if a message is not successfully sent. I mean, if a customer sends an e-mail to our domain and misspells the SMTP address of one of our sales people ? You want an NDR to go back to them so hopefully they realize their mistake.Spamcop.net even says to use SPF for checking the e-mail origin? Well, I use SPF. But only block e-mails where the sending domain provides an SPF record and the authentication fails. I am not going to block emails coming into our domain just because a sending domain may not have SPF setup for their domain? I mean, I cant force them to provide and SPF record, even though it is recommended.SpamCop.net users should either stop relying on their services or either use SpamCop.net in a ?weighted? approach for determining SPAM.Any way ? I had to gripe about this poor decision on SpamCop?s behalf and would like to get your opinions?Regards,Chris Wall - MCSE + MessagingNAM Exchange AdministratorT (919) 460.3236F (919) 468.4889Global KnowledgeLEARNING. To Make a Difference.
