Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Im running an email server as part of a research study. We have subjects connecting and sending email from a variety of clients.

One function of this server is to communicate with said subjects - this means sending out regular emails. In the last few weeks it seems that nearly all of these emails are being marked as spam by the big providers (gmail, hotmail, yahoo, etc). As a result I'm looking into how to mark these messages as safe.

One suggestion that comes up from searching is to use DKIM. I can set this up and try it but it certainly won't be in place for all of our existing clients. If I implement it will it block all emails from our server that aren't setup to use it?

FWIW: Postfix, CentOS 6.4 x64

share|improve this question
up vote 1 down vote accepted

You can configure DKIM and only sign a subset of your messages. This probably won't help solve your actual problem though.

There is no completely reliable method to ensure your messages are not marked as spam. Every provider has different detection and filtering mechanisms.

There are several things you can do to reduce the chance that your messages will be marked as spam.

  1. Messages should be properly formed. This means that all necessary headers are present and that none of those headers are malformed or contain invalid data.
  2. Systems sending mail should have rDNS configured for the sending IP address.
  3. Properly configured SPF records should be maintained for your domains.
  4. Sign messages with DKIM.
  5. Handle abuse complaints promptly.
  6. Proactively monitor RBLs and resolve any listing issues.
  7. Sanity check destination addresses before sending messages. Prune mailing lists of e-mail addresses that consistently bounce.
  8. Don't actually send unsolicited e-mail. This really could be a separate discussion entirely. Require confirmation of subscription requests. Provide a link to easily unsubscribe within your bulk messages.
share|improve this answer
None of these emails are from subscriptions or unsolicited. It's all communications from the study that these people are part of. – ethrbunny Feb 19 '14 at 19:50
@ethrbunny I wasn't trying to imply that you were sending unsolicited mail. I was simply rounding out the list for the benefit of anyone that comes across this question in the future. – Mark Sturgill Feb 19 '14 at 19:53
Absolutely! I appreciate the comprehensive list. I'm just trying to think of what rules I'm breaking. – ethrbunny Feb 19 '14 at 20:00
@ethrbunny Missing/Incorrect SPF and rDNS are the two most common issues. There are many web-based tools that can check your sending IP address against the most popular RBLs as well. – Mark Sturgill Feb 19 '14 at 20:05

If a mail server doesn't support DKIM then it will simply be ignored.

share|improve this answer

Your Answer


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.