On Tue, Apr 14, 2015 at 02:26:35PM -0500, Andrew Noonan wrote:
> I suspect that this is the root of the problem. As Viktor kindly
> pointed out, there are probably other things that should be tweaked,
> but as this was pretty much all a/b delay, and not c/d delay.
Don't confuse cause and effect. High "c+d" delays *cause* high
"b" delays as mail accumulates in the active queue and later even
high "a" delays when the active queue becomes full, and mail sits
in "incoming".
> I'll
> probably walk the concurrency value (in the right file) up a little
> from a default of 20 and make sure that Google doesn't get mad with
> me. Is it considered good form to report back, or is that list spam?
Come back if you have further questions. I used to discourage
vacuous "it worked" follow-ups, but they are not taboo, your call.
My take is that if you don't have something material to add to the
analysis that led to the solution, no need to post, but if the
actual solution involved additional insights feel free to summarize.
--
Viktor.
