Hi Glen,
Thats certainly possible. But just to be clear, I wasn't referring to
the GL blocking delays, but the outbound sending retry frequency.
The facts are real: GL systems exist and growing every day. So even if a
system is not interested in implementing it on the receiving side, there
is a delayed delivery impact due to the increasing odds of hitting a
remote GL compliant receiver. A shorted 2nd attempt will minimize this
impact.
I personally think your GL block of 25 minutes is too high for most
systems with a higher expectation of business or customer transactions.
But thats ok? It works for you. :-)
Hmmmm, I wonder if there could be an algoritm or heuristic to make the
block time variable too based on who is the sender? But I do think the
beauty of GL is that it treats all anonymous senders the same.
--
HLS
Glenn Anderson wrote:
I recommend a 5 min retry on the 2nd attempt, with a fall back to
their normal interval after that.
My MTA has a default graylisting delay of 25 minutes. 25 minutes was
chosen because it wouldn't unnecessarily delay for hosts that use the
recommended 30 minute retry, and because at the time I saw a significant
number of what appeared to be bots that were retrying 5 times 5 minutes
apart (~20 minutes between the first and last retry).
Glenn.
--
Sincerely
Hector Santos, CTO
http://www.santronics.com
http://santronics.blogspot.com