[courier-users] missing MX record

[SZÉPE Viktor]

Hello Sam!

RFC 5321 states in
https://tools.ietf.org/html/rfc5321#section-5

The lookup first attempts to locate an MX record associated with the name.
... If an empty list of MXs is returned,
the address is treated as if it was associated with an implicit MX
RR, with a preference of 0, pointing to that host.

Were you a ware of that?
I think it is very unusual and dangerous.

Do modern MTA-s - including Courier - implement that?



SZÉPE Viktor
https://github.com/szepeviktor/debian-server-tools/blob/master/CV.md

[Matus UHLAR - fantomas]

Post by SZÉPE Viktor
RFC 5321 states in
https://tools.ietf.org/html/rfc5321#section-5

The lookup first attempts to locate an MX record associated with the name.
... If an empty list of MXs is returned,
the address is treated as if it was associated with an implicit MX
RR, with a preference of 0, pointing to that host.

Were you a ware of that?
I think it is very unusual and dangerous.
Do modern MTA-s - including Courier - implement that?

This behaviour was described in rfc 821 and 2821.
AFAIK all MTAs implement this behaviour since MX records were implemented.

What and why exactly sounds unusual and dangerous to you?

[SZÉPE Viktor]

Post by Matus UHLAR - fantomas

Post by SZÉPE Viktor
RFC 5321 states in
https://tools.ietf.org/html/rfc5321#section-5

The lookup first attempts to locate an MX record associated with the name.
... If an empty list of MXs is returned,
the address is treated as if it was associated with an implicit MX
RR, with a preference of 0, pointing to that host.

Were you a ware of that?
I think it is very unusual and dangerous.
Do modern MTA-s - including Courier - implement that?

This behaviour was described in rfc 821 and 2821.
AFAIK all MTAs implement this behaviour since MX records were implemented.
What and why exactly sounds unusual and dangerous to you?

I think it gives us no means to stop emails for a domain.
I thought removing the MX record and not listening on port 25 is enough.

This way anyone my send an email to a mailserver-less sub/domain.


SZÉPE Viktor
https://github.com/szepeviktor/debian-server-tools/blob/master/CV.md

[Matus UHLAR - fantomas]

Post by SZÉPE Viktor

Post by Matus UHLAR - fantomas

RFC 5321 states in https://tools.ietf.org/html/rfc5321#section-5

The lookup first attempts to locate an MX record associated with the name.
... If an empty list of MXs is returned,
the address is treated as if it was associated with an implicit MX
RR, with a preference of 0, pointing to that host.

Were you a ware of that?
I think it is very unusual and dangerous.
Do modern MTA-s - including Courier - implement that?

This behaviour was described in rfc 821 and 2821.
AFAIK all MTAs implement this behaviour since MX records were implemented.
What and why exactly sounds unusual and dangerous to you?

I think it gives us no means to stop emails for a domain.
I thought removing the MX record and not listening on port 25 is enough.
This way anyone my send an email to a mailserver-less sub/domain.

This mechanism was created when MX records were introduced, to support
host/domains without them.

This is how things should be done - creating new standard and define how
backwards compatibility should be implemented.

Read rfc 7505 that tries to implement mechanism to archieve that as a new
measurement, and don't blame us for implementign something that has existed
even before MX and was never dropped since.