On Oct 22, 2008, at 11:59 AM, Graham Leggett wrote:
Jim Jagielski wrote:
Except we don't enforce an upper limit at all... That's
the point :)
I don't see any point in enforcing an upper limit - if I want a 1MB
buffer size, and I have enough RAM, there is no real reason I can
see why I shouldn't have it.
If the ajp protocol limits this to 64k this is also fine, as long as
it is documented correctly.
I think this is really an issue of the docs needing to be clarified
and corrected, rather than the need for a code change.
Except that with the current code, one cannot set it to be less than
8192... I also don't feel strongly about enforcing an upper limit, but
I don't think we should behind-the-scenes adjust so it's a multiple
of 512/1024. I suggested 512 as a lower limit to compliment the rec
buffer.
