John Levine <[email protected]> wrote:

>>The one complication I can think of is that if there is the potential 
>>for getting multiple messages, there's no way to tell if you are going
>
>>to get more.
>
>If that were really a problem (it's not clear to me why it would be),
>it wouldn't be hard to invent a line like Group-Count: with a hint of
>how many related reports were sent to the same target at about the
>same time.
>
>My impression is that you're more likely to want to correlate multiple
>reports about similar failures to different messages rather than
>different failures to one message.

In my experience it's generally about trying to associate failure types and 
sources. Having information about multiple authentication types can be helpful 
in this.

Having all the facts about one message arrive together feels right from an 
system perspective. It's either arrived or not and there's no need to consider 
partial availability of data. OTOH, it may just be a minor corner case.

I suspect that all the information in one mail makes this RFC a bit more 
complicated, but it will make some types of implementations that make use of 
this data less complex and subject to race conditions.

Scott K
_______________________________________________
marf mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/marf

Reply via email to