Frank Ellermann wrote:
Radu Hociung wrote:
Here's how their SPF record maps out:
Fascinating, you found a very different interpretation of the
processing limits, you _accumulate_ MX queries. Bye, Frank
Hi Frank,
Yes, I think the way limits should be counted can be improved. Perhaps I
haven't found the best way, and I am open to suggestions, but I think
enforcing limits is more controllable if there are accumulated counts
for all the different lookup types, and also an overall total counter.
MX, A and TXT records are about equally expensive to perform. PTR
records are more expensive because usually they take longer to resolve,
and often time out. I can't substantiate it, but my experience is such.
So since not all lookups are equally expensive, should they not be
subjected to different limits?
Do you see a problem with accumulating query counts across the entire
SPF record evaluation process ?
Thanks,
Radu.