spf-discuss
[Top] [All Lists]

RE: DNS Load Summary

2005-03-22 10:29:00
...... Original Message .......
On Tue, 22 Mar 2005 11:03:03 -0600 Andy Bakun 
<spf(_at_)leave-it-to-grace(_dot_)com> 
wrote:
On Tue, 2005-03-22 at 08:34, Scott Kitterman wrote:
- no disagreement that all clients should use the same limit.

I would say that senders should be guaranteed that all clients will 
process
at least X mechanisms.  If someone wants to do more work than that, I 
don't
think the spec should prohibit that.

Do more work to what end?  If someone decides to do more work to still
look for a Pass, then the results will be unpredictable.  I'm fine with
saying "SPF evaluators can do as much work as they want, but the final
result should be arrived at before limit X is reached.  If limit X is
reached with no definite result, then result code Y is returned. 
Clients can continue processing the record beyond limit X, if they want
to waste their own resources."

Does it make sense to continue processing the record after you get a
Pass?


No.  I'm just trying to avoid a future problem where some bean counter 
says, "Your implementation isn't RFC compliant because it does X + 1 and 
the RFC says X."

Anyway this is more likely to come up on a fail than a pass.

Scott K



<Prev in Thread] Current Thread [Next in Thread>