On Wed, May 8, 2013 at 5:40 AM, Daniel Stenberg wrote:
> On Tue, 7 May 2013, Alex Loukissas wrote:
>
> Yes rather strange behavior, although it's not very frequent. Do you
>> perhaps have a hunch on anything I might have missed in the way I'm using
>> curl_mult
On Tue, May 7, 2013 at 3:00 PM, Daniel Stenberg wrote:
> On Tue, 7 May 2013, Alex Loukissas wrote:
>
> I'm using the curl_multi interface to make parallel GET requests on a set
>> of URLs. I've noticed that at times the call to curl_multi_wait may block
>> inde
s a perhaps relevant blog
post<http://daniel.haxx.se/blog/2012/10/10/wsapoll-is-broken/>,
which could be related.
By the way, this was seen on Win7 and my app is using a libcurl 7.28.1,
statically linked.
Thanks!
--
Alex Loukissas
Member of Technical Staff - Maginatic
--
> List admin: http://cool.haxx.se/list/listinfo/curl-library
> Etiquette: http://curl.haxx.se/mail/etiquette.html
>
--
Alex Loukissas
Member of Technical Staff - Maginatics, Inc.
www.maginatics.com
---
List admin: http://cool.haxx.se/list/listinfo/curl-library
Etiquette: http://curl.haxx.se/mail/etiquette.html
nd Windows Phone 8.
>
>
> --
>
> / daniel.haxx.se
>
> --**--**---
> List admin:
> http://cool.haxx.se/list/**listinfo/curl-library<http://cool.haxx.se/list/listinfo/curl-library>
> Etiquette:
> http://curl.haxx.se/mail/**etiquette.html<http://curl.haxx.se/mail/etiquette
On Wed, Sep 19, 2012 at 2:07 AM, Daniel Stenberg wrote:
> On Tue, 18 Sep 2012, Alex Loukissas wrote:
>
> These disappear when I define CURL_DISABLE_POP3 and CURL_DISABLE_SMTP.
>> Should we perhaps add these two when user defines HTTP_ONLY?
>>
>
> Yes indeed, and ther
On Wed, Sep 19, 2012 at 2:07 AM, Daniel Stenberg wrote:
> On Tue, 18 Sep 2012, Alex Loukissas wrote:
>
>> These disappear when I define CURL_DISABLE_POP3 and CURL_DISABLE_SMTP.
>> Should we perhaps add these two when user defines HTTP_ONLY?
>
>
> Yes indeed, and there a
l_create_ntlm_type3_message
1>libcurl.lib(pop3.obj) : error LNK2001: unresolved external symbol
Curl_sasl_cleanup
1>libcurl.lib(smtp.obj) : error LNK2001: unresolved external symbol
Curl_sasl_cleanup
These disappear when I define CURL_DISABLE_POP3 and CURL_DISABLE_SMTP.
Should we perhaps add
hanks!
Alex
On Tue, Mar 13, 2012 at 8:54 PM, Alex Loukissas wrote:
> Hi Guenter,
>
> I've actually figured out what was the issue, it was user error :-)
> Here's the snippet of code I used:
>
> int writePubKey(const unsigned char* input, int length, FILE* fp) {
>
px
Thanks for your help!
Alex
On Tue, Mar 13, 2012 at 6:15 PM, Guenter wrote:
>
> Hi Alex,
> Am 13.03.2012 19:22, schrieb Alex Loukissas:
>
>> I've been trying to remove the requirement of having to bundle a
>> certificate file, as described here
>> [http://
ren't valid for common use. If this works, it would
nicely eliminate the pain of having to bundle up certificate files with
apps using cURL and I'd be happy to share the complete code.
Thanks!
--
Alex Loukissas
Member of Technical Sta
Hello,
I have an app that links statically against libcurl, which itself (libcurl)
is built with openssl support. My app is cross-compiled on Linux with
mingw-64 and run on Win7-64. The problem I'm facing is the following:
Unless I use the CURLOPT_CAINFO option and hard-code a certificate path,
t
> I will run this experiment and come back with the results.
>
Initial runs show no network inbound traffic when the download thread
is hung on poll.
My hunch is still that the issue is due to the fact that 2 threads are
trying to do a GET on the same URL. I will try to reproduce with a
small pro
> Doesn't this simply "hang" for 1000 milliseconds and then it continues?
No, it hangs for a couple minutes. My test scenario is a download of a
large number of objects, with total size of approx 15MB. One thread
(the one that's hung makes a single call to readData (with the list of
all objects),
I have been debugging further, having rebuilt libcurl with debug info.
Here's the backtrace from within the hung thread:
(gdb) bt
#0 0x00844ed3 in poll ()
#1 0x00561674 in Curl_socket_ready (readfd=20, writefd=-1,
timeout_ms=1000) at select.c:215
#2 0x0058b384 in Transfe
> libcurl has no locking but then libcurl does nothing by itself that needs to
> get locked. You however said that your function that is called by multiple
> threads do:
>
> CURL* handle = get a handle from a pool of handles,
>
> If this pool isn't per-thread, you need to protect it, and hence I
Since it's been rather quiet, I'd like to restate my question for clarification:
Is there a possibility of a race for 2 threads that use different curl
easy handles that are trying to access the same URL?
Thanks
Alex
---
List admin:
Digging a little deeper, I discovered what may be happening:
All objects (and their corresponding URLs) are of fixed size, 64KB.
Let's assume that the URL is http://www.example.com/object-1-2-3. The
case is that 2 threads are trying to fetch the same URL by calling the
readData function. One threa
On Thu, Nov 3, 2011 at 2:56 AM, Daniel Stenberg wrote:
> So there's also some kind of mutex or semaphore action going on there I
> assume?
So far the readData function has no locking. From what I understand,
this seems to be the root of the problem.
> This is too little information. Does data a
Hi everyone,
I'm having an issue similar to
http://curl.haxx.se/mail/lib-2008-09/0171.html, with a twist. Short
description:
- Multi-threaded C++ application, with a readData() function that may
be called from multiple threads.
- The readData function does (in pseudocode) the following:
CURL*
On Thu, Sep 15, 2011 at 2:23 PM, Daniel Stenberg wrote:
> On Thu, 15 Sep 2011, Alex Loukissas wrote:
>
>>> Why can't you just use the existing void * argument?
>>
>> Correct me if I'm wrong, but doesn't this contain the data read by curl to
>>
On Thu, Sep 15, 2011 at 2:12 PM, Daniel Stenberg wrote:
> On Thu, 15 Sep 2011, Alex Loukissas wrote:
>
>> My question is whether libcurl supports/will support a different kind
>> of signature for the callback function, one with an extra (void *)
>> argument, where we can
Hello,
I'm experiencing a difficulty using write callbacks and I'd like to
ask how/whether is possible to do what I have to do. I outline the
scenario briefly below:
* The callback I want to use is a public member function of class A.
The reasoning behind this is that I'd like to update some stat
Great thanks!
Alex
On Tue, Sep 6, 2011 at 2:42 PM, Daniel Stenberg wrote:
> On Tue, 6 Sep 2011, Alex Loukissas wrote:
>
>> I wanted a clarification on the comment "Future versions of libcurl are
>> likely to not have any built-in progress meter at all." in the l
Hello,
I wanted a clarification on the comment "Future versions of libcurl
are likely to not have any built-in progress meter at all." in the
libcurl API manual. Is all support for progress callbacks dropped from
future releases of libcurl?
Thanks
Alex
On Thu, Aug 11, 2011 at 2:07 PM, Daniel Stenberg wrote:
> On Thu, 11 Aug 2011, Alex Loukissas wrote:
>
>>> The multi interface does all the transfers you add to it in parallell.
>>
>> So does it do it in a multi-threaded fashion (probably not, since I didn't
&
On Thu, Aug 11, 2011 at 1:14 PM, Daniel Stenberg wrote:
> On Thu, 11 Aug 2011, Alex Loukissas wrote:
>
>> Is there a benefit over what I'm doing now (i.e. looping through the URIs
>> and issuing a curl_easy_perform) versus having a number of handles in a
>> c
torage access, or network that is the
> bottleneck, and the characteristics of the bottleneck (ie spikes vs
> sustained bottleneck) should help you formulate a better solution if you are
> really looking to maximize your perf.
>
> On Thu, Aug 11, 2011 at 11:29 AM, Alex Loukissas
>
On Thu, Aug 11, 2011 at 10:18 AM, Daniel Stenberg wrote:
> On Thu, 11 Aug 2011, Alex Loukissas wrote:
>
>>> In many situations you won't gain any performance by doing parallell
>>> uploads to a single server, and you'll get a simpler implementation by doing
&
Hi Daniel,
Thanks for the reply. Couple comments inline.
On Thu, Aug 11, 2011 at 3:01 AM, Daniel Stenberg wrote:
> On Wed, 10 Aug 2011, Alex Loukissas wrote:
>
>> I am looking to refactor this code now, so that I get a speedup by
>> parallelism (the order that these requ
The scenario is the following:
I want to perform many GET and PUT requests on a single HTTP server
for many files (e.g. downloading/uploading large photo albums), where
the only thing changing between each request is the name of the file.
The current code I have, takes the URI I wish to GET/PUT an
31 matches
Mail list logo
