>
> Appears to be a firewall issue. I made a dummy PHP script that opens
> connection and does simple post without curl. So, it hangs like the curl
> test. But for some reason only on multipart/form-data posts. Disabling
> firewall allows multipart post.. so I think there's not a curl issue here.
On Mon, Mar 7, 2016 at 3:42 PM, Waitman Gobble wrote:
>
>
> On Mon, Mar 7, 2016 at 2:56 PM, Ray Satiro via curl-library <
> curl-library@cool.haxx.se> wrote:
>
>> On 3/7/2016 3:10 AM, Waitman Gobble wrote:
>>
>>> Thanks for the replies, I will read through them and figure it
On Mon, Mar 7, 2016 at 2:56 PM, Ray Satiro via curl-library <
curl-library@cool.haxx.se> wrote:
> On 3/7/2016 3:10 AM, Waitman Gobble wrote:
>
>> Thanks for the replies, I will read through them and figure it out. I
>> really appreciate it.
>> I didn't post the entire file.
>>
>> One thing I
On Mon, 7 Mar 2016, Inca R wrote:
> Possibly we introduce an option that makes libcurl use HTTP/2 at once over
> http://
Is the "option" ready yet ?
Nope, no work has been done for that as far as I know. It's open for grabs!
--
/ daniel.haxx.se
On 3/7/2016 3:10 AM, Waitman Gobble wrote:
Thanks for the replies, I will read through them and figure it out. I
really appreciate it.
I didn't post the entire file.
One thing I think the problem is in lib/http.c
else if(data->set.postfields)
expectsend =
Hi;
This is from: https://curl.haxx.se/docs/http2.html
"
...
Possibly we introduce an option that makes libcurl use HTTP/2 at once over
http://
"
Is the "option" ready yet ?
We have http/2 already implemented using nghttp2 to proxy between http/1.1
and http/2. nghttp2 supports http/1.1 upgrade
On Mon, Mar 7, 2016 at 3:30 PM, Pierre Brico wrote:
>
> Hi Isaac,
>
> To my knowledge, the multi interface allows just you to manipulate several
> easy handles at a time (asynchronously) but doesn't give access to the
> underlying connection to the server. Am I wrong ?
On Mon, 7 Mar 2016, Pierre Brico wrote:
So my problem is that libcurl manages its own connections pool. In this
pool, there are connections to the web server 1 and the web server 2 but
each virtual user has its HTTP session on one server (there is no
replication between web servers). So, for
On Mon, Mar 7, 2016 at 7:17 AM, Waitman Gobble wrote:
> On Mon, Mar 7, 2016 at 6:59 AM, Waitman Gobble wrote:
>> On Mon, Mar 7, 2016 at 1:54 AM, Daniel Stenberg wrote:
>>> On Mon, 7 Mar 2016, Waitman Gobble wrote:
>>>
I'm still
On Mon, 7 Mar 2016, Waitman Gobble wrote:
At the moment it appears to be stuck in curl_multi_wait() in lib/multi.c
What did it send and receive before it got stuck?
--
/ daniel.haxx.se
---
List admin:
On Mon, Mar 7, 2016 at 6:59 AM, Waitman Gobble wrote:
> On Mon, Mar 7, 2016 at 1:54 AM, Daniel Stenberg wrote:
>> On Mon, 7 Mar 2016, Waitman Gobble wrote:
>>
>>> I'm still seeing it hang, tomorrow I'll attach to debugger and see what is
>>> sticking.
>>
>>
On Mon, Mar 7, 2016 at 1:54 AM, Daniel Stenberg wrote:
> On Mon, 7 Mar 2016, Waitman Gobble wrote:
>
>> I'm still seeing it hang, tomorrow I'll attach to debugger and see what is
>> sticking.
>
>
> If you show us a full program that reproduces the hang we might be able to
> tell
Hi Isaac,
To my knowledge, the multi interface allows just you to manipulate several
easy handles at a time (asynchronously) but doesn't give access to the
underlying connection to the server. Am I wrong ?
Kind regards,
Pierre
On Mon, Mar 7, 2016 at 9:19 AM, Isaac Boukris
On Wed, 2 Mar 2016, Boutin Maƫl wrote:
I'm currently using libcurl to send data over a nginx local server using
HTTP POST and transfer-encoding chunked.
My application is fed with buffers which i have to send as soon as i get
them. These buffers may belong to various endpoints on the nginx
Hi Daniel,
I've identified the issue I have. The web application I'm benchmarking is
located behind a loadbalancer which forwards the requests to 2 different
web servers. When a new connection is established between the client and
one of the 2 servers, the loadbalancer adds a new cookie (which
On Mon, 7 Mar 2016, Waitman Gobble wrote:
I'm still seeing it hang, tomorrow I'll attach to debugger and see what is
sticking.
If you show us a full program that reproduces the hang we might be able to
tell you why.
--
/ daniel.haxx.se
>
> I take it that pipelining is required to trigger this, isn't it?
That is correct.
>
> To me it sounds like a decent way to fix this problem, so please by all means
> show us your take!
>
--- curl-7.47.1.orig/lib/http.h 2016-02-02 15:02:44.0 -0800
+++ curl-7.47.1/lib/http.h
On Mon, Mar 7, 2016 at 1:26 AM, Waitman Gobble wrote:
> On Mon, Mar 7, 2016 at 1:15 AM, Daniel Stenberg wrote:
>> On Mon, 7 Mar 2016, Waitman Gobble wrote:
>>
>>> I've tried with and without Content-Length header.
>>> Without adding the header it was sending
On Mon, Mar 7, 2016 at 1:15 AM, Daniel Stenberg wrote:
> On Mon, 7 Mar 2016, Waitman Gobble wrote:
>
>> I've tried with and without Content-Length header.
>> Without adding the header it was sending 172 bytes, which I believe is
>> due to strlen()
>
>
> Test 1531 tests pretty much
On Mon, 7 Mar 2016, Waitman Gobble wrote:
I've tried with and without Content-Length header.
Without adding the header it was sending 172 bytes, which I believe is
due to strlen()
Test 1531 tests pretty much exactly this and it works fine.
What libcurl version are you using?
--
/
On Mon, Mar 7, 2016 at 12:53 AM, Daniel Stenberg wrote:
> On Mon, 7 Mar 2016, Waitman Gobble wrote:
>
>> Sorry, original post doesn't show but i did add it in
>> long post_size;
>> post_size = ftell(fp);
>>
>> curl_easy_setopt(conn, CURLOPT_POSTFIELDSIZE, post_size);
>
>
> But you
On Mon, 7 Mar 2016, Isaac Boukris wrote:
I currently have a similar requirement and I'm testing the approach of
coupling each easy handle with its own multi handle (using
curl_multi_socket_action).
Oh right, that should be doable.
It works ok so far, though I wonder how well it would scale
On Mon, 7 Mar 2016, Waitman Gobble wrote:
Sorry, original post doesn't show but i did add it in
long post_size;
post_size = ftell(fp);
curl_easy_setopt(conn, CURLOPT_POSTFIELDSIZE, post_size);
But you overrode that value yourself in the request:
std::string cl = "Content-Length: " +
On Mon, Mar 7, 2016 at 12:31 AM, Daniel Stenberg wrote:
> On Mon, 7 Mar 2016, Daniel Stenberg wrote:
>
>>> One thing I think the problem is in lib/http.c
>>>
>>> else if(data->set.postfields)
>>>expectsend = (curl_off_t)strlen(data->set.postfields);
>>
>>
>> Hm. That
On Mon, 7 Mar 2016, Waitman Gobble wrote:
I pulled curl from git, and changed http.c to check for
data->set.postfieldsize , getting ready to test i'll let you know.
That's not your problem. Read the rest of the function where 'expectsend' is
used and you'll notice that isn't used for your
On Mon, Mar 7, 2016 at 12:25 AM, Daniel Stenberg wrote:
> On Mon, 7 Mar 2016, Waitman Gobble wrote:
>
>> One thing I think the problem is in lib/http.c
>>
>> else if(data->set.postfields)
>>expectsend = (curl_off_t)strlen(data->set.postfields);
>
>
> Hm. That doesn't
On Mon, 7 Mar 2016, Daniel Stenberg wrote:
One thing I think the problem is in lib/http.c
else if(data->set.postfields)
expectsend = (curl_off_t)strlen(data->set.postfields);
Hm. That doesn't look right...
That said, your use case won't trigger any problem because of that as
On Mon, 7 Mar 2016, Waitman Gobble wrote:
One thing I think the problem is in lib/http.c
else if(data->set.postfields)
expectsend = (curl_off_t)strlen(data->set.postfields);
Hm. That doesn't look right...
--
/ daniel.haxx.se
On Mon, Mar 7, 2016 at 12:10 AM, Waitman Gobble wrote:
> On Sun, Mar 6, 2016 at 11:43 PM, Ray Satiro via curl-library
> wrote:
>> On 3/7/2016 1:15 AM, Waitman Gobble wrote:
>>>
>>> When I set Content-Type header to multipart/form-data and add data
Hi,
On Mon, Mar 7, 2016 at 1:15 AM, Daniel Stenberg wrote:
> On Sun, 6 Mar 2016, Pierre Brico wrote:
>
>> So my question is: is it possible to link one connection to one CURL
>> handle (no pool usage)?
>
>
> In theory it should be fairly easy to have each easy handle use its own
Daniel Stenberg wrote:
> On Fri, 4 Mar 2016, Gisle Vanem wrote:
>
>> I'm built libcurl using BoringSSL (MSVC-2015) but it crashes for a very good
>> reason. Here is the call-stack (slightly
>> edited):
>
> David Benjamin in the BoringSSL team produced a patch for this problem and
> it'd be
On Sun, Mar 6, 2016 at 11:43 PM, Ray Satiro via curl-library
wrote:
> On 3/7/2016 1:15 AM, Waitman Gobble wrote:
>>
>> When I set Content-Type header to multipart/form-data and add data to
>> CURLOPT_POSTFIELDS it seems to hang. I thought maybe proxy issue but
>> tried
32 matches
Mail list logo