Bug#687065: maxproc is quite correct

2012-09-10 Thread Jonas Smedegaard
On 12-09-10 at 04:37pm, Thomas Goirand wrote: > On 09/10/2012 03:06 PM, Jonas Smedegaard wrote: > >My point is that setting DKIMproxy to 5 and Postfix to 10 is wrong: > I'm sorry, but it doesn't make sense. The documentation in the > README.Debian tells to do this: > smtp inet n - n

Bug#687065: maxproc is quite correct

2012-09-10 Thread Thomas Goirand
On 09/10/2012 03:06 PM, Jonas Smedegaard wrote: My point is not that 10 is better than 5. My point is that setting DKIMproxy to 5 and Postfix to 10 is wrong: it works but quite inefficiently - if you don't care about queues being handled correct then why care about those numbers at all? On a re

Bug#687065: maxproc is quite correct

2012-09-10 Thread Jonas Smedegaard
On 12-09-10 at 02:01pm, Thomas Goirand wrote: > During a discussion on upstream mailing list, I remember that one > person did some tests to check what was the optimum value for maxproc. > If I remember well, he told that 5 was the ideal value for a single > core CPU. > > So I wonder: how many

Bug#687065: maxproc is quite correct

2012-09-09 Thread Thomas Goirand
Hi Jonas, After giving it a day, and thinking a bit more about your bug entry, I have some more to add. During a discussion on upstream mailing list, I remember that one person did some tests to check what was the optimum value for maxproc. If I remember well, he told that 5 was the ideal va