On 7/1/2016 3:57 AM, Jgh wrote:
> Sorry, yes; you are correct about the endpass.
>
> So the next suggestion is to run a test with debug enabled and examine the cutthrough processing sequence. -bh and -d options.
Thanks for the suggestion. I didn't realize there was an option to test
as if from a remote IP. Nice! I did it with -bhc, not -bh, though, since
-bh doesn't do the callout verification part. And I verified that the
callout part is happening (and that Barracuda verifies non-existent
recipients, which is what's causing most of the problem in the first
place!) but I don't see anything that shows cutthrough happening.
I am correct in thinking that cutthrough *should* be happening, right?
And that my system should not accept the message from the sender until
the DATA has been acked by the destination system, right? Have any
examples of a correct, working configuration I can compare to?
--
Rob Szarka
http://szarka.org/
