Normally I’m fine with the filters discourse has setup to avoid posting data for maillings, however how would I best approach allow discourse to create a post anyway with a “Precedence: bulk” header set for a specific domain?
The reason I’m asking, one of our resellers is sending all mails with mandril and has a Precedence: bulk header set which doesn’t get posted as a discourse app. We obviously do want to deal with these mails, anyone have an idea on what best to do here?
I will be adding a new setting today to deal with that specific issue. Instead of rejecting “auto generated” emails, they will be marked as such but will be processed like any other emails.
Cool! But hopefully this doesn’t mean the return of “out of office” replies. 
It’ll be a site setting. But we’re experiencing about 50% false positives and want to feel the pain in order to understand it more.
So to confirm this is related to Google Apps as well,
User sends an email to an email address and the mail gets flagged as “Email::Receiver::AutoGeneratedEmailError”
Also is there a feature not send reject emails for the auto-generated stuff? I think it causes confusion when an out of office email gets caught.
You should configure bounce email, cause that should eliminate the majority
Not entirely sure I follow what you’re suggesting…
See
Most software sends out of office replies to the return path, some does not
Here’s the new block_auto_generated_emails site setting (defaults to true) 
https://github.com/discourse/discourse/commit/4960b621101845017f7bd4146f7f8d155a8b41e4
