On Wed, Mar 9, 2011 at 9:12 AM, Philip Martin
<philip.mar...@wandisco.com> wrote:
> Joe Orton <jor...@redhat.com> writes:
>
>> On Wed, Mar 09, 2011 at 08:50:37AM -0800, Justin Erenkrantz wrote:
>>> On Wed, Mar 9, 2011 at 8:28 AM, Philip Martin
>>> <philip.mar...@wandisco.com> wrote:
>>> > 0.000054 epoll_wait(3, {{EPOLLOUT, {u32=30583944, u64=30583944}}},
>>> > 16, 3600000000) = 1 <0.000011>
>>>
>>> As best as I can tell, that strace isn't matching your earlier
>>> description. The last epoll_wait is taking 0.000011 seconds? --
>>
>> Smells like Nagle to me - looks like serf is turning Nagle *on* not
>> off... ("nodelay off" == "delay on") - try this?
>>
>> Index: outgoing.c
>> ===================================================================
>> --- outgoing.c (revision 1440)
>> +++ outgoing.c (working copy)
>> @@ -201,7 +201,7 @@
>>
>> /* Disable Nagle's algorithm */
>> if ((status = apr_socket_opt_set(skt,
>> - APR_TCP_NODELAY, 0)) !=
>> APR_SUCCESS)
>> + APR_TCP_NODELAY, 1)) !=
>> APR_SUCCESS)
>> return status;
>>
>> /* Configured. Store it into the connection now. */
>
> Yes, that's it. Serf is now comparable to neon.
This patch is now applied to trunk and 0.7.x. Thanks! -- justin
