On Sun, Aug 14, 2011 at 04:15:10PM +0300, about bus wrote:
> I've checked this, domain name resolution does not take much time.
> Issue somewhere between end of evhttp_socket_cb and start of evhttp_read_cb
> functions calls.
>
> 2011.08.14 16:52:27 [warn] === evhttp_socket_cb: fd: 131
> 2011.08.1
An autoconf setting (to use largefile or not) would be a great start,
and I like the idea of defaulting to 64-bit, even though it will
probably break some people who don't compile that way but use off_t in
their code. From what I can tell, a 64-bit interface to the OS is more
general, and then
Looks good, and I agree the BSD code wasn't broken. :)
On 8/15/2011 10:39 AM, Nick Mathewson wrote:
On Thu, Aug 11, 2011 at 5:16 PM, Michael Herf wrote:
In addition to my note about 64-bit offsets earlier:
Solaris sendfile seems to fail when sending moderately large (<1GB) files.
Not a 32/64
On Thu, Aug 11, 2011 at 5:16 PM, Michael Herf wrote:
> In addition to my note about 64-bit offsets earlier:
>
> Solaris sendfile seems to fail when sending moderately large (<1GB) files.
> Not a 32/64 problem, but a buffer problem.
> Anyone else ever try this? It is definitely broken in http-serve
On Thu, Aug 11, 2011 at 2:06 PM, Michael Herf wrote:
> I'm compiling libevent on Solaris.
> ev_off_t appears to be #define'd as off_t on UNIX machines ("#define
> ev_off_t off_t" in util.h).
> However, on this machine, off_t is first defined in *stdio.h*!
> Additionally, my app uses the following
On Mon, Aug 15, 2011 at 19:26, Nick Mathewson wrote:
>> Also, feel free to include evdns-example.c as sample of proper evdns
>> cleanup if you find it useful. I think, that cleanup is not trivial.
>
> Hm. I'd like to have something about dns usage in the samples code,
> but this one really only co
On Mon, Aug 15, 2011 at 7:10 AM, Leonid Evdokimov wrote:
> I think, I've found couple of memory leaks in libevent debugging code:
> the event was not properly unregistered from debug hashtable.
Good catch here; thanks!
> I run evdns-example using $(valgrind --leak-check=full
> --show-reachable=y
Merged; thanks!
***
To unsubscribe, send an e-mail to majord...@freehaven.net with
unsubscribe libevent-usersin the body.
I think, I've found couple of memory leaks in libevent debugging code:
the event was not properly unregistered from debug hashtable.
I run evdns-example using $(valgrind --leak-check=full
--show-reachable=yes), I'm attaching three valgrind logs:
evdns-example.no-0004.no-0005.log - without 0004-*.
--
WBRBW, Leonid Evdokimov
xmpp:l...@darkk.net.ru && http://darkk.net.ru
tel:+79816800702 && tel:+79050965222
From 0bb4044523158b86ee12314d86ee4a3d39d29fd1 Mon Sep 17 00:00:00 2001
From: Leonid Evdokimov
Date: Thu, 11 Aug 2011 03:06:07 +0400
Subject: [PATCH 3/5] Another docstring fix.
---
includ
On Mon, Aug 15, 2011 at 12:30 PM, Leonid Evdokimov wrote:
> On Fri, Aug 12, 2011 at 22:48, about bus wrote:
> > So, if everything ok with network, where libevent or maybe OS (FreeBSD
> 7.2)
> > holds request for 30 / 75 seconds?
>
> Do you use ACCEPT_FILTER in your http server? I think it can cre
On Fri, Aug 12, 2011 at 22:48, about bus wrote:
> So, if everything ok with network, where libevent or maybe OS (FreeBSD 7.2)
> holds request for 30 / 75 seconds?
Do you use ACCEPT_FILTER in your http server? I think it can create
such delay in case of some nasty bug.
--
WBRBW, Leonid Evdokimov
12 matches
Mail list logo
