Home | Sitemap | Recent Changes | Login

SPF Logo

Sender Policy Framework

Frank Ellermann/OAQ

Once Answered Questions

For obscure reasons folks using the Contact form often use "topic: The SPF Website" when they actually mean "topic: Help Request". I'd be willing to approve anything that is no spam for the SPF Webmaster list, but folks on this list hate it when it's clearly off topic. I'm not inclined to start a free "play by mail" support, we have already the SPF Help list and the great SPF Support team for this task.

Nevertheless sometimes I answer off topic questions in private mail instead of discarding the moderation request, on this page I'll try to collect what turned out to be an interesting communication with a sentient being.

"?all" vs. "~all"

What do you recommend, using question mark "?" or "~" in my spf record?

If you are sure that you don't want SPF FAIL protection you can use "?all" at the end.

If you want to test the water for a limited time (some weeks) test "~all", and if all is as you want it definitely go for "-all", the full SPF FAIL protection.

IMO "?all" is for cowards, but for "-all" you need to know what you tell your users if they get an unexpected FAIL for their good mail - that can happen if simplified receivers test SPF at the wrong place, not at their border.

You can see a big collection of domains using "-all" at <http://spf-all.com/> - obviously they survived it, and some have millions of users (gmx.net).

Don't use "~all" forever, make up your mind. With "-all" receivers are supposed to reject FAIL, that is a clear situation. With "~all" receivers do "something else", it is hard to diagnose if "something" is not what you want.

With "?all" you have no protection, but you can still gain a good reputation for a PASS, that was your original plan.

That was a productive discussion, it helped to find an oddity with the SenderID wizard and to submit a related feature request for the SPF validator.

Edit text of this page | View other revisions
Last edited 2011-11-05 19:11 (UTC) by Julian Mehnle (diff)