Daniel Stenberg ac022b2e30 Christopher R. Palmer reported a problem with HTTP-POSTing using "anyauth"
that picks NTLM. Thanks to David Byron letting me test NTLM against his
servers, I could quickly repeat and fix the problem. It turned out to be:

When libcurl POSTs without knowing/using an authentication and it gets back a
list of types from which it picks NTLM, it needs to either continue sending
its data if it keeps the connection alive, or not send the data but close the
connection. Then do the first step in the NTLM auth. libcurl didn't send the
data nor close the connection but simply read the response-body and then sent
the first negotiation step. Which then failed miserably of course. The fixed
version forces a connection if there is more than 2000 bytes left to send.
2005-02-16 14:31:23 +00:00
..
2004-08-10 10:43:41 +00:00
2004-12-15 01:38:25 +00:00
2004-11-29 12:11:46 +00:00
2004-12-11 18:47:22 +00:00
2004-12-11 18:47:22 +00:00
2004-12-11 18:47:22 +00:00
2004-12-17 19:57:50 +00:00
2004-12-11 18:47:22 +00:00
2004-08-16 13:24:01 +00:00
2002-06-14 09:36:09 +00:00
2004-05-24 11:57:34 +00:00
2004-01-07 09:19:33 +00:00
2004-01-07 09:19:33 +00:00
2004-11-11 16:34:24 +00:00
2004-01-07 09:19:33 +00:00
2004-08-16 11:09:01 +00:00
2004-12-08 23:02:55 +00:00
2004-08-16 11:09:30 +00:00
2004-06-09 01:15:03 +00:00
2005-01-14 13:43:29 +00:00
2004-12-17 12:28:04 +00:00
2004-12-11 18:47:22 +00:00
2004-12-15 01:38:25 +00:00
2004-12-15 01:38:25 +00:00
2004-12-15 01:38:25 +00:00
2004-01-07 09:19:33 +00:00
2004-06-10 21:20:15 +00:00
2004-04-06 07:59:11 +00:00
2004-04-30 08:51:19 +00:00
2005-01-14 13:43:29 +00:00
2003-07-22 10:00:37 +00:00
2005-02-10 07:45:08 +00:00
2004-10-11 17:26:24 +00:00
2004-01-07 09:19:33 +00:00
2004-12-17 18:33:09 +00:00
2004-12-17 18:32:41 +00:00
2004-12-17 17:49:10 +00:00
2005-02-09 22:47:57 +00:00

$Id$
                                  _   _ ____  _     
                              ___| | | |  _ \| |    
                             / __| | | | |_) | |    
                            | (__| |_| |  _ <| |___ 
                             \___|\___/|_| \_\_____|

             How To Track Down Suspected Memory Leaks in libcurl
             ===================================================

Single-threaded

  Please note that this memory leak system is not adjusted to work in more
  than one thread. If you want/need to use it in a multi-threaded app. Please
  adjust accordingly.


Build

  Rebuild libcurl with -DCURLDEBUG (usually, rerunning configure with
  --enable-debug fixes this). 'make clean' first, then 'make' so that all
  files actually are rebuilt properly. It will also make sense to build
  libcurl with the debug option (usually -g to the compiler) so that debugging
  it will be easier if you actually do find a leak in the library.

  This will create a library that has memory debugging enabled.

Modify Your Application

  Add a line in your application code:

       curl_memdebug("filename");

  This will make the malloc debug system output a full trace of all resource
  using functions to the given file name. Make sure you rebuild your program
  and that you link with the same libcurl you built for this purpose as
  described above.

Run Your Application

  Run your program as usual. Watch the specified memory trace file grow.

  Make your program exit and use the proper libcurl cleanup functions etc. So
  that all non-leaks are returned/freed properly.

Analyze the Flow

  Use the tests/memanalyze.pl perl script to analyze the memdump file:

    tests/memanalyze.pl < memdump

  This now outputs a report on what resources that were allocated but never
  freed etc. This report is very fine for posting to the list!

  If this doesn't produce any output, no leak was detected in libcurl. Then
  the leak is mostly likely to be in your code.