Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

Background:
I'm about to launch a service which offers email forwarding for a large number of domains.

Say you want people to be able to contact you using you@somesite.com, all emails sent to that address would be forwarded to your@realaddress.com

We will have 3 servers with their own IPs, the servers will run Postfix and all domains will have abuse@ and postmaster@ addresses which will forward to our admins.

I don't want our IPs to get blocked, all we're doing is forwarding the received messages, but I don't really want to enable spam blocking on the forwarded mail (It's our users, not ours, it should be up to their onbound email provider to detect spam and put it in a 'spam' folder, the only option we have is to delete spam messages but if we start getting it wrong and people start getting upset that we're not delivering their emails we could get in trouble with them)

One option is frequently (maybe once a month) get new IPs for the servers so they should be clean (We can get them for about $2/year per IP so no huge cost to us, but we end up with a lot of 'soiled' ips lying around which our hosting company won't take back). Are there any other suggestions for what we can do so we don't get blocked by email providers?

Edit:
We'll be forwarding between 15-20k emails a day to ~10k different accounts. We're currently using OpenSRS' email forwarding service but it's too expensive now we've got to this level of users

share|improve this question
1  
I'd have to imagine this is one of the areas where utilizing spf records will be a big benefit as well. –  thinice Jun 20 '11 at 22:29

2 Answers 2

up vote 6 down vote accepted

Basically, the better you comply with the relevant RFCs the less likely you will be blacklisted. There are no guarantees, but there are a number of things that will get you low trust ratings and/or blacklisted.

  • Invalid or missing rDNS verification data.
  • Mail server not using its FQDN in HELO/EHLO command.
  • Using dynamic IP addresses.
  • Otherwise, forging or appearing to forge your server's identity.
  • Not having an SPF record allowing the server to send email.
  • Forwarding email for addresses with SPF records which don't allow you to forward.
  • Sending email to harvested addresses (your clients control this).
  • Forwarding spam from your clients.

I have done a number of posting on Electronic Mail which may help you. In particular you may want to look at my posting on Detecting Server Forgery.

share|improve this answer

No guarantees. Even MessageLabs gets blacklisted, and that's not the only time it happened to them.

Will you send it "from" your company domain? If you send "from" the sender's address and they have SPF on their domain, that will make the messages get flagged as high-spam-probability.

share|improve this answer
    
Currently they'll come from the original sender, but I didn't know whether it would be better to use a proper address (say forwarder@ourdomain.com) in the From field and set a reply-to, not sure how hard that is in Postfix but that's a different question –  sam Jun 21 '11 at 1:04
    
Use the Sender: header to get around the SPF problem. –  yakatz Aug 5 '11 at 5:20

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.