Am 30.10.20 um 08:57 schrieb Atro Tossavainen via mailop:
> Why does Google bounce after accepting a message? At Google's scale,
> the potential to become the world's biggest spammer simply through
> backscatter is enormous.
>
Sadly, doing the correct thing isn't always technically possible.

The mail server ingress needs to decide whether to accept a mail message based 
on criteria that are relatively fast to
evaluate. DNSBL lookups, SPF-checks etc can be done fast. Part of this is 
applying heuristics do reject mails with
apparently fake senders (that's where SPF can be useful, if you ignore the 
issue with forwarders).

Once mail has been accepted into the system, more expensive checks may still be 
advisable (malware checks etc.). If you
find that a message should not have been accepted in the first place you're in 
a slightly inconvenient situation. You may

  * silently discard it and hope that nobody notices since it was actually sent 
by spamware which doesn't follow up and
    it won't be missed by the recipient. Although this may often be the 
"correct" option there may be legal problems
    with this approach.
  * forward it to the recipient with some "spam/malware/whatever" tag or 
destination folder. This would probably keep
    you on the safe side legally while exposing inexperienced or gullible users 
to the risk of picking up malware from
    their junk folder.
  * bounce it back to the sender which you assume is valid (since you did some 
due diligence on the ingress server to
    reject fakes). That way, you don't drop the message, and you don't bother 
your customer. Let the sender sort it out
    if they actually exist.

Each option has its pros and cons. I prefer not to bounce back messages but 
others may choose a different policy.

I suspect that Google is good enough at validating senders that becoming the 
world's biggest spammer through backscatter
is not an issue (they are already letting hacked mailboxes or spammer-owned 
throwaway accounts spam quite a bit, while
I've yet to see significant backscatter from them).

Cheers,
Hans-Martin

_______________________________________________
mailop mailing list
mailop@mailop.org
https://list.mailop.org/listinfo/mailop

Reply via email to