On Wed, 27 Mar 2002, Rogerio Saran wrote:

> I am writing a wrapper around libcurl (7.9.5) to make it suitable to be
> called through libwww-perl (LWP). The goal is to replace the current LPW
> https handler with a higher performance one.

Sounds like an interesting project.

[snip]

> > POST /GRIPNET/gracgi.EXE HTTP/1.1^M
> Host: bankline.itau.com.br^M
> Pragma: no-cache^M
> Accept: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, */*^M
> User-Agent: Mozilla/4.0 (compatible; MSIE 4.01; Windows 98)^M
> Content-Length: 91^M
> Content-Type: application/x-www-form-urlencoded^M
> ^M
> ...post data...
>
> >>>>> oh no!
>
> * Closing connection #0

Do you happen to know why it decides that this connection needs to be closed?

> Organox::Protocol::LibCurl::request: Curl::easy::perform result 0
> Organox::Protocol::LibCurl::request: request 8.773619
> Organox::Protocol::LibCurl::request: read 26547 bytes
>
> >>>>> the response code is "100 continue"...
>
> Organox::Protocol::LibCurl::build_response: HTTP/1.1 100 Continue
> Organox::Protocol::LibCurl::build_response: Date: Wed, 27 Mar 2002
> 13:58:56 GMT
> Organox::Protocol::LibCurl::__ANON__: Date: Wed, 27 Mar 2002 13:58:56 GMT
> Organox::Protocol::LibCurl::__ANON__: Server: Microsoft-IIS/4.0
> Organox::Protocol::LibCurl::request: final 0.002665
> LWP::Protocol::collect: read 26458 bytes
>
> Anybody wonder why libcurl is closing the connection even after receiving a
> "continue" response code?

Yes, I wonder why.

AFAICR, the 100-response code just eats that and then awaits the next "real"
header so it should close the connection because of this.

Are you sure it isn't the proxy or server that closes this connection?

-- 
    Daniel Stenberg -- curl groks URLs -- http://curl.haxx.se/