Hello,
I'm just making it through RFCs about SMTP protcol, and I'd have a quick question. It seems like most of the SMTP servers nowadays have AUTH verb implemented, which is quite understandable.

So, if I want to send an email, I'm going to contact my SMTP server, to which I have to authenticate using AUTH. The SMTP server then sends my message to another SMTP server, from where the message can be read by target user (using some other protocol - not important now).

I tried to simulate incoming message on my SMTP server - and I failed, getting 550 status code - Relay not permitted.

So my question is, what's the difference between me and some other unknown SMTP server trying to contact mine to transfer a mail?


Thanks,

Skyer

Yes, just about all SMTP relays are set up to use authentication. Open relays are primarily used by spammers, and are quickly blacklisted by RBL maintainers.

It would seem that your server doesn't permit relaying at all. The difference between you and a random server on the Internet trying to use your server as a relay, is that you should be able to authenticate with a valid username and password, and/or your connection should come from a trusted IP range.

You need to set up authentication and relaying in your server. I can't be more specific since you haven't told us which mail server you're using.

Thank you for your answer. It seems like I haven't described my question enough.

The server I am using is not 'mine' - I don't own it, it's just email service provided when you buy a domain. (So by mine, I rather meant the server on my side. I have an email account there, but not system's user account).

The authentication problem is clear, what I don't understand is, how is it possible for some other, totally foreign SMTP server to contact mine and transfer message, when I can't do it myself.

IP is not a way to go, as far as I know there's no global database of SMTP servers. (Or is there?)


Thanks,

Skyer

There is a global database. For every (mail-enabled) domain, there is an MX record pointing to one or more FQDNs of mail servers that will accept mail for those domains. How else would any mail get to the right recipient? Mail couldn't very well just bounce around the 'net at random until it happend to reach the right server.

I don't know exactly what you tried to do, but if you tried to send a mail using SMTP to the registered mail server for your domain, with your own e-mail address as the recipient, it should have worked.

However, most mail servers filter mail against RBLs, and if your test originated from a blacklisted IP address (such as a dynamic IP handed out by an ISP), that could account for your mail being rejected.

2 members found this post helpful.

Ah,
so that's the magic behind it

It looks quite simple when you've got it explained, my bad that I haven't looked at DNS deeply so far.

(Of course email couldn't arrive without DNS, but I thought that the record is simply for all types of query, and does not differ for mail servers http servers, etc.)

Thanks for explanation, problem solved.


Skyer