cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1390
Views
0
Helpful
4
Replies

AsyncOS violating RFC 2034

Donald Nash
Level 3
Level 3

This is mainly addressed to the IronPort employees who hang out here.

I found this thread on a discussion board while searching for something else. The bottom line is that AsyncOS does not follow RFC 2034 in its use of enhanced SMTP status codes because it precedes them with a "#", like this:

550 #5.1.0 Address rejected.

instead of this:

550 5.1.0 Address rejected.

This prevents SMTP clients which implement RFC 2034 from noticing the enhanced status code. AsyncOS doesn't advertise the ENHANCEDSTATUSCODES capability, so it isn't actually breaking the rules. But if IronPort is going to go to the effort of putting those codes in there, then why not follow RFC 2034?

Thanks.

4 Replies 4

Heya,

That is kinda right, but we never announce after the EHLO that we do follow the advanced status codes.

Cheers,

Mark

Donald Nash
Level 3
Level 3

[quote:1903d3ddb0="Mark [CSE]"]That is kinda right, but we never announce after the EHLO that we do follow the advanced status codes.
Yes I know. That's what I said:

AsyncOS doesn't advertise the ENHANCEDSTATUSCODES capability, so it isn't actually breaking the rules.

My question is, since you're putting the enhanced codes in there anyway, why not do it in a way that conforms to RFC 2034 so that conformant clients can use them? No one recognizes Dan Bernstein's sharp style, so those codes aren't really doing any good except to human readers. But human readers have the explanatory text. The enhanced codes are meant to be machine readable, so it's useless to put them in there in a way that negates this purpose.

Thanks,

bfayne_ironport
Level 1
Level 1

You could call tech support and ask them to open a formal feature request. Product Management does look at them.

If they can determine that removing the # sign won't really break anything, it might not be that hard to implement.

Donald Nash
Level 3
Level 3

I will file a feature request, and I can't imagine how removing the "#" will hurt anything. But I am curious to know why they chose to do it that way in the first place.