ChatGPT解决这个技术问题 Extra ChatGPT

What's the difference between Sender, From and Return-Path?

What's the difference between an email Sender, From and Return-Path value?

Example: I have a contact form where the user can input their email, would this be assigned to sender, from or return-path?

I had a quick search on the StackOverflow and couldn't find anything useful.


S
Shawn D.

So, over SMTP when a message is submitted, the SMTP envelope (sender, recipients, etc.) is different from the actual data of the message.

The Sender header is used to identify in the message who submitted it. This is usually the same as the From header, which is who the message is from. However, it can differ in some cases where a mail agent is sending messages on behalf of someone else.

The Return-Path header is used to indicate to the recipient (or receiving MTA) where non-delivery receipts are to be sent.

For example, take a server that allows users to send mail from a web page. So, sender@yourcompany.com types in a message and submits it. The server then sends the message to its recipient with From set to sender@yourcompany.com. The actual SMTP submission uses different credentials, something like mailagent@mywebmail.com. So, the sender header is set to mailagent@mywebmail.com, to indicate the From header doesn't indicate who actually submitted the message.

In this case, if the message cannot be sent, it's probably better for the agent to receive the non-delivery report, and so Return-Path would also be set to mailagent@mywebmail.com so that any delivery reports go to it instead of the sender.

If you are doing just that, a form submission to send e-mail, then this is probably a direct parallel with how you'd set the headers.


As well, you don't have to set everything. i.e. if you leave out sender & return path, they go to the From address. If you leave out return path, NDRs go to the sender, I think.
... Which is an attractive nuisance for mail bombers. Don't do that!
Do I understand this right? When talking about a web form for submitting an e-mail, the Sender is the person who submitted the web form and the From is the server that sent out the e-mail? Or is it the other way round?
Imagine some VIP that has an assistant managing their mailbox. If the assistant is writing email on behalf of the VIP, the assistant is the Sender, but the message is From the VIP. This is what happens when you see email described as "From Assistant on behalf of VIP"
@ShawnD., What if there is no Return-Path. Does it default to the Sender then?
C
Community

The official RFC which defines this specification could be found here:

https://www.rfc-editor.org/rfc/rfc4021#section-2.1.2 (look at paragraph 2.1.2. and the following)

2.1.2. Header Field: From Description: Mailbox of message author [...] Related information: Specifies the author(s) of the message; that is, the mailbox(es) of the person(s) or system(s) responsible for the writing of the message. Defined as standard by RFC 822. 2.1.3. Header Field: Sender Description: Mailbox of message sender [...] Related information: Specifies the mailbox of the agent responsible for the actual transmission of the message. Defined as standard by RFC 822. 2.1.22. Header Field: Return-Path Description: Message return path [...] Related information: Return path for message response diagnostics. See also RFC 2821 [17]. Defined as standard by RFC 822.


Thanks for giving the official RFC link. It's really useful if someone ask, "Based on what?"
This other answer (from 2011) claims that the method indicated here causes gmail to flag emails as spam. I wonder if that's still the case today.
Updated in RFC 5322 tools.ietf.org/html/rfc5322#section-3.6 . Can someone please tell the SMTP RFC people that it would help if the Sender field has to match the From address used in the SMTP handshake process if it is to be used.
C
Community

A minor update to this: a sender should never set the Return-Path: header. There's no such thing as a Return-Path: header for a message in transit. That header is set by the MTA that makes final delivery, and is generally set to the value of the 5321.From unless the local system needs some kind of quirky routing.

It's a common misunderstanding because users rarely see an email without a Return-Path: header in their mailboxes. This is because they always see delivered messages, but an MTA should never see a Return-Path: header on a message in transit. See https://www.rfc-editor.org/rfc/rfc5321#section-4.4


A sender using an email client would not set it, but a "sender" writing a script that sends emails may set it scriptually, thus I think its misleading to say a sender should never set it.
Unfortunately, Chilinut is actually inaccurate. A Return-Path: header on a message in transit will be discarded, the MDA (mail deliver agent) that performs final delivery will set the Return-Path: header to match the value of the 5321.From (envelope-from) carried by the message. This is because the envelope is lost when the message is delivered, so the Return-Path: header records what the envelope-from was as the MDA received the message.
The Return-Path: header reflects the envelope-from, or RFC5321.From address. The From: header reflects the header-from, or RFC5322.From address.
It's getting to be semantics, the important thing (as above) is that you cannot set a Return-Path: header when sending a message. If ones does, it will be discarded in transit and later set to the value of the RFC5321.From or envelope-from by the MDA that makes final delivery of the message. Basically the Return-Path: header records what the envelope-from had been, as the envelope is discarded on delivery.
The return path in the envelope of a message is the same as the envelope from. However, the Return-Path: header is part of the body of the message and is set by the MDA.