Hector Santos wrote:
Send a 421 per RFC 2821 section 3.9.
One problem with 421 is that it normally means something else: I
believe that 421 normally means that the SMTP server is shutting down
for some reason. But a server which times-out a connection is not
shutting down -- it is still operating and open for business. So for
such a server to send a 421 is a little bit of a lie, and may be
misinterpreted by a client.
... This is about idle timeouts (long inactivity) right?
Yes. This is about when a server is sitting and waiting for the next
command from the client, for some number of minutes (This could be a
denial of service attack). And it is also about too slow transmission
of data after the DATA command. If for instance the client sends only
one packet every two minutes, and each packet contains only one
additional byte of data, that could be cause for the server to timeout
after some amount of time (A denial of service attack could also take
this form).
What I am saying if you are going to hang up due to long activity, the only
thing that is useful is the response text. The code itself is meaningless
since the SMTP client would not know what to do with it.
Thank you for what you imply, but do not say directly, here. You
imply that the RFC does not tell what the server should say in this
case. You also imply that there is no standard or common practice for
what the server should say in this case, because otherwise the client
would know what to do.
Can you say -- forthrightly with confidence -- what you imply? that
there is no formalized or standard message for the SMTP server to send
just before it exercises its option to timeout because of long
inactivity or too slow data transmission?
Thank you,
Rich