Jeremy Zapata | 1 May 17:51 2015

Setting MAC option

I am using libssh2 compiled with OpenSSL and libcurl (OpenSSL v1.0.2 and libssh2 v1.4.3, libcurl 7.22.0) and I use it to test uploading/downloading files over sftp.

I have limited experience with ssh and sftp but on Linux there are additional options that can be specified with the sftp command such as the MAC (Message Authentication Code) like this:

sftp -oCiphers=3des-cbc -oMACs=hmac-md5 myuser <at> myhost

 

Is anyone familiar with the MAC option and how it can be set for libssh2 (hopefully there is a way through libcurl)?

 

It was hard to find a whole lot of information on the MAC option, but this is something I found: http://www.computerhope.com/unix/sftp.htm.

 

Thanks!

 

Jeremy Zapata | Software Engineer in Test
globalscape | +1 (210) 308-8267 ext 5176 | NYSE MKT:GSB

*This communication, including attachments, is for the exclusive use of the addressee and may contain proprietary, confidential or privileged information. If you are not the intended recipient, any use, copying, disclosure, dissemination or distribution is strictly prohibited. If you are not the intended recipient, please notify the sender immediately by return email and delete this communication and destroy all copies.

 

 

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
赵志毅 | 29 Apr 04:59 2015

how to get libssh2-devel source code

hi,

I am trying to cross compile zabbix on mips platform. It needs libssh2 and libssh2-devel. I can download libssh2, but there is no way to download libssh2-devel.

Where is available for libssh2-devel?

cheers

armstrong zhao

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
libssh2 Trac | 14 Apr 11:29 2015
Picon

#302: libssh2_session_handshake bug?

#302: libssh2_session_handshake bug?
--------------------+--------------------
 Reporter:  5m0k3   |       Owner:
     Type:  defect  |      Status:  new
 Priority:  normal  |   Milestone:  1.5.0
Component:  crypto  |     Version:  1.5.0
 Keywords:          |  Blocked By:
   Blocks:          |
--------------------+--------------------
 Hello, I was working on a password auditing tool using libssh2 and
 encountered the following issue when using with pthread.

 GDB output:

 [Thread debugging using libthread_db enabled]
 Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
 Core was generated by `./dm domains.txt user pass 500'.
 Program terminated with signal SIGSEGV, Segmentation fault.
 #0  0x00007ff0eaa3bbdc in ?? () from /lib/x86_64-linux-
 gnu/libcrypto.so.1.0.0
 (gdb) bt
 #0  0x00007ff0eaa3bbdc in ?? () from /lib/x86_64-linux-
 gnu/libcrypto.so.1.0.0
 #1  0xca62c1d6ca62c1d6 in ?? ()
 #2  0xca62c1d6ca62c1d6 in ?? ()
 #3  0xca62c1d6ca62c1d6 in ?? ()
 #4  0xca62c1d6ca62c1d6 in ?? ()
 #5  0xca62c1d6ca62c1d6 in ?? ()
 #6  0xca62c1d6ca62c1d6 in ?? ()
 #7  0xca62c1d6ca62c1d6 in ?? ()
 #8  0xca62c1d6ca62c1d6 in ?? ()
 #9  0x00007ff0eada8810 in ?? () from /lib/x86_64-linux-
 gnu/libcrypto.so.1.0.0
 #10 0x00007feeb000e400 in ?? ()
 #11 0x000000000000000f in ?? ()
 #12 0x00007ff0eaa3811f in SHA1_Update () from /lib/x86_64-linux-
 gnu/libcrypto.so.1.0.0
 #13 0x00007ff0eaab76ef in ?? () from /lib/x86_64-linux-
 gnu/libcrypto.so.1.0.0
 #14 0x00007ff0eaa6d72c in ?? () from /lib/x86_64-linux-
 gnu/libcrypto.so.1.0.0
 #15 0x00007ff0eb399f8a in diffie_hellman_sha1 () from
 /usr/lib/libssh2.so.1
 #16 0x00007ff0eb39c197 in
 kex_method_diffie_hellman_group14_sha1_key_exchange () from
 /usr/lib/libssh2.so.1
 #17 0x00007ff0eb39e1ad in _libssh2_kex_exchange () from
 /usr/lib/libssh2.so.1
 #18 0x00007ff0eb3a6c6e in session_startup () from /usr/lib/libssh2.so.1
 #19 0x00007ff0eb3a6efd in libssh2_session_handshake () from
 /usr/lib/libssh2.so.1
 #20 0x0000000000402023 in verhost (hostname=0x25f1700 "be-still.info",
 username=0x2560400 "be-still", password=0x25c1590 "be-still") at dm.c:248
 #21 0x00000000004021e2 in worker_thread (worker=0x1f2) at dm.c:287
 #22 0x00007ff0eb179182 in start_thread (arg=0x7fefb2fc1700) at
 pthread_create.c:312
 #23 0x00007ff0eaea647d in clone () at
 ../sysdeps/unix/sysv/linux/x86_64/clone.S:111
 (gdb) frame 20
 #20 0x0000000000402023 in verhost (hostname=0x25f1700 "be-still.info",
 username=0x2560400 "be-still", password=0x25c1590 "be-still") at dm.c:248
 248             rc = libssh2_session_handshake(session, sockfd);
 (gdb)

 Here is the code for function causing problems:

 static int verhost(char *hostname, char *username, char *password) {
         if(hostname == NULL) return -1;
         if(username == NULL) return -1;

         int sockfd, rc;
         long arg;

         struct addrinfo hints, *res;
         int errcode;

         memset (&hints, 0, sizeof (hints));
         hints.ai_family = PF_UNSPEC;
         hints.ai_socktype = SOCK_STREAM;
         hints.ai_flags |= AI_CANONNAME;

         errcode = getaddrinfo (hostname, "22", &hints, &res);
         if (errcode != 0) return -1;

         sockfd = socket(AF_INET, SOCK_STREAM, 0);

         arg = fcntl(sockfd, F_GETFL, NULL);
         arg |= O_NONBLOCK;
         fcntl(sockfd, F_SETFL, arg);

         if (sockfd < 0) goto shutend;

         if (connect(sockfd, res->ai_addr, res->ai_addrlen) < 0) {
                 if (errno == EINPROGRESS) {
                         fd_set write_fds;
                         memset(&write_fds, 0, sizeof(write_fds));
                         FD_ZERO(&write_fds);
                         FD_SET(sockfd, &write_fds);
                         struct timeval tv;
                         int valopt;
                         tv.tv_sec = 5;
                         tv.tv_usec = 0;
                         if (select(sockfd+1, NULL, &write_fds, NULL, &tv)
 > 0) {
                                 socklen_t lon;
                                 lon = sizeof(int);
                                 getsockopt(sockfd, SOL_SOCKET, SO_ERROR,
 (void*)(&valopt), &lon);
                                 if (valopt) goto shutend;
                         }
                         else goto shutend;
                 }
                 else goto shutend;
         }

         LIBSSH2_SESSION *session = libssh2_session_init();
         if(!session) goto shutend;

         libssh2_session_set_timeout(session, 10000);
         if(libssh2_session_get_timeout(session) > 10000) goto shutnow;

         rc = libssh2_session_handshake(session, sockfd);
         if (rc) goto shutnow;

         rc = libssh2_userauth_password(session, username, password);
         if (rc) printf("[!] Weak Login: %s - %s/%s\n", hostname, username,
 password);
         else libssh2_session_disconnect(session, "Normal Shutdown");

 shutnow:
         libssh2_session_free(session);

 shutend:
         close(sockfd);
         freeaddrinfo(res);
         return -1;
 }

 Let me know if you need anything else or if I am using libssh2 in a wrong
 manner. Thanks.

--

-- 
Ticket URL: <https://trac.libssh2.org/ticket/302>
libssh2 <https://trac.libssh2.org/>
C library for writing portable SSH2 clients

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
libssh2 Trac | 10 Apr 12:03 2015
Picon

#301: random crashes during multi thread

#301: random crashes during multi thread
---------------------+--------------------
 Reporter:  alexmol  |       Owner:
     Type:  defect   |      Status:  new
 Priority:  normal   |   Milestone:  1.5.0
Component:  crypto   |     Version:  1.5.0
 Keywords:           |  Blocked By:
   Blocks:           |
---------------------+--------------------
 It seems like random crashes may occur in RAND_bytes function. In my app I
 use libssh with 20 parallel threads, so RAND_bytes called quite agressive
 from diff threads. After I guard it with critical section - everything
 seems ok. I cannot really understand the RAND_bytes code, because it looks
 scary. Seems like it is not thread safe. Openssl version for me is 0.9.8zf

--

-- 
Ticket URL: <https://trac.libssh2.org/ticket/301>
libssh2 <https://trac.libssh2.org/>
C library for writing portable SSH2 clients

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
libssh2 Trac | 3 Apr 18:54 2015
Picon

Re: #168: Memory leak in libssh2_session_startup

#168: Memory leak in libssh2_session_startup
-----------------------+--------------------
  Reporter:  jcollins  |      Owner:
      Type:  defect    |     Status:  closed
  Priority:  normal    |  Milestone:
 Component:  API       |    Version:  1.2.4
Resolution:  fixed     |   Keywords:
Blocked By:            |     Blocks:
-----------------------+--------------------
Changes (by alamaison):

 * status:  new => closed
 * resolution:   => fixed

Comment:

 Thanks.  Pushed
 https://github.com/libssh2/libssh2/commit/09c5e59933daf67b833f34b8c388766abc038483.

--

-- 
Ticket URL: <https://trac.libssh2.org/ticket/168#comment:6>
libssh2 <https://trac.libssh2.org/>
C library for writing portable SSH2 clients

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
libssh2 Trac | 3 Apr 14:14 2015
Picon

Re: #168: Memory leak in libssh2_session_startup

#168: Memory leak in libssh2_session_startup
-----------------------+-------------------
  Reporter:  jcollins  |      Owner:
      Type:  defect    |     Status:  new
  Priority:  normal    |  Milestone:
 Component:  API       |    Version:  1.2.4
Resolution:            |   Keywords:
Blocked By:            |     Blocks:
-----------------------+-------------------

Comment (by alamaison):

 Replying to [comment:4 ssbssa]:
 > Like this, or did you mean more?

 Could you resubmit your patch with a different filename?  There's a bug in
 Trac that prevents the moderator accepting your submission, because the
 patch name already exists in this ticket.

--

-- 
Ticket URL: <https://trac.libssh2.org/ticket/168#comment:5>
libssh2 <https://trac.libssh2.org/>
C library for writing portable SSH2 clients

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
libssh2 Trac | 3 Apr 14:12 2015
Picon

Re: #168: Memory leak in libssh2_session_startup

#168: Memory leak in libssh2_session_startup
-----------------------+-------------------
  Reporter:  jcollins  |      Owner:
      Type:  defect    |     Status:  new
  Priority:  normal    |  Milestone:
 Component:  API       |    Version:  1.2.4
Resolution:            |   Keywords:
Blocked By:            |     Blocks:
-----------------------+-------------------

Comment (by ssbssa):

 Like this, or did you mean more?

--

-- 
Ticket URL: <https://trac.libssh2.org/ticket/168#comment:4>
libssh2 <https://trac.libssh2.org/>
C library for writing portable SSH2 clients

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
libssh2 Trac | 3 Apr 14:08 2015
Picon

#300: Memory leak in kex.c

#300: Memory leak in kex.c
------------------------------+--------------------
 Reporter:  ElectronShepherd  |       Owner:
     Type:  defect            |      Status:  new
 Priority:  normal            |   Milestone:  1.5.0
Component:  crypto            |     Version:  1.5.0
 Keywords:  memory leak       |  Blocked By:
   Blocks:                    |
------------------------------+--------------------
 I'm picking up a memory leak in kex.c, line 205. The code is:

 {{{
         session->server_hostkey =
             LIBSSH2_ALLOC(session, session->server_hostkey_len);

 }}}

 For some execution paths, session->server_hostkey is already allocated, so
 when the contents of the pointer are replaced, there's a leak.

 Oddly, I found it while SSHing into a Synology device, and it's fully
 repeatable. What's strange is that I don't get the leak when connecting to
 a "proper" Linux install such as Debian 7.

 For now, my workaround is to add

 {{{
             if (session->server_hostkey) {
                     LIBSSH2_FREE(session, session->server_hostkey);
                 }

 }}}

 just above the leaking call, but that doesn't seem like the most reliable
 solution to me, since it seems like the code is reusing a session which
 hasn't been cleared properly, and there's probably more cleanup that
 should be done against the session.

--

-- 
Ticket URL: <https://trac.libssh2.org/ticket/300>
libssh2 <https://trac.libssh2.org/>
C library for writing portable SSH2 clients

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
libssh2 Trac | 3 Apr 02:42 2015
Picon

Re: #168: Memory leak in libssh2_session_startup

#168: Memory leak in libssh2_session_startup
-----------------------+-------------------
  Reporter:  jcollins  |      Owner:
      Type:  defect    |     Status:  new
  Priority:  normal    |  Milestone:
 Component:  API       |    Version:  1.2.4
Resolution:            |   Keywords:
Blocked By:            |     Blocks:
-----------------------+-------------------

Comment (by alamaison):

 The fix looks good.  Could you add some comments to the patch so that, in
 a year's time, people aren't scratching their heads trying to work out why
 the gcrypt version uses NULL.

--

-- 
Ticket URL: <https://trac.libssh2.org/ticket/168#comment:3>
libssh2 <https://trac.libssh2.org/>
C library for writing portable SSH2 clients

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
libssh2 Trac | 1 Apr 20:09 2015
Picon

Re: #168: Memory leak in libssh2_session_startup

#168: Memory leak in libssh2_session_startup
-----------------------+-------------------
  Reporter:  jcollins  |      Owner:
      Type:  defect    |     Status:  new
  Priority:  normal    |  Milestone:
 Component:  API       |    Version:  1.2.4
Resolution:            |   Keywords:
Blocked By:            |     Blocks:
-----------------------+-------------------

Comment (by ssbssa):

 The problem here is that gcry_mpi_scan() doesn't use the bignum provided
 by the argument, instead it creates a new one and overrides the one
 created by gcry_mpi_new() before.

 I've attached a possible solution.

--

-- 
Ticket URL: <https://trac.libssh2.org/ticket/168#comment:2>
libssh2 <https://trac.libssh2.org/>
C library for writing portable SSH2 clients

_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
Markus Doppelbauer | 1 Apr 13:28 2015
Picon

Memory leaks

Hello,

First, thanks a lot for the new release. Maybe there are some
memory leaks:

a) #298: Clean openssl-tables via "EVP_cleanup()":
 https://www.openssl.org/docs/crypto/OpenSSL_add_all_algorithms.html

 Replace:

 #define libssh2_crypto_init() OpenSSL_add_all_algorithms()
 #define libssh2_crypto_exit()

 With:

 #define libssh2_crypto_init() OpenSSL_add_all_algorithms()
#define libssh2_crypto_exit() EVP_cleanup();

b) #299: _libssh2_rsa_new() in openssl.c:69 is not free'd()
I have attached the valgrind-output.

Thanks a lot
Markus
==3020== Memcheck, a memory error detector
==3020== Copyright (C) 2002-2013, and GNU GPL'd, by Julian Seward et al.
==3020== Using Valgrind-3.10.0.SVN and LibVEX; rerun with -h for copyright info
==3020== Command: bin/Debug/update
==3020== 
==3020== 
==3020== HEAP SUMMARY:
==3020==     in use at exit: 416 bytes in 6 blocks
==3020==   total heap usage: 5,825 allocs, 5,819 frees, 1,593,012 bytes allocated
==3020== 
==3020== 24 bytes in 1 blocks are still reachable in loss record 1 of 6
==3020==    at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==3020==    by 0x80DED72: CRYPTO_malloc (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E00E5: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E0A8A: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x8147926: RSA_new_method (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x5CADD56: _libssh2_rsa_new (openssl.c:69)
==3020==    by 0x5C99785: hostkey_method_ssh_rsa_init (hostkey.c:95)
==3020==    by 0x5C99E4E: diffie_hellman_sha1 (kex.c:267)
==3020==    by 0x5C9B2A6: kex_method_diffie_hellman_group14_sha1_key_exchange (kex.c:799)
==3020==    by 0x5C9BD89: _libssh2_kex_exchange (kex.c:1770)
==3020==    by 0x5CA2756: libssh2_session_handshake (session.c:723)
==3020==    by 0x912119: ssh2_connect(String const&, unsigned long, String const&, String const&, String
const&) (ssh.cpp:1395)
==3020== 
==3020== 24 bytes in 1 blocks are still reachable in loss record 2 of 6
==3020==    at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==3020==    by 0x80DED72: CRYPTO_malloc (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x8164C7A: lh_insert (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E0113: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E0A8A: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x8147926: RSA_new_method (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x5CADD56: _libssh2_rsa_new (openssl.c:69)
==3020==    by 0x5C99785: hostkey_method_ssh_rsa_init (hostkey.c:95)
==3020==    by 0x5C99E4E: diffie_hellman_sha1 (kex.c:267)
==3020==    by 0x5C9B2A6: kex_method_diffie_hellman_group14_sha1_key_exchange (kex.c:799)
==3020==    by 0x5C9BD89: _libssh2_kex_exchange (kex.c:1770)
==3020==    by 0x5CA2756: libssh2_session_handshake (session.c:723)
==3020== 
==3020== 32 bytes in 1 blocks are still reachable in loss record 3 of 6
==3020==    at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==3020==    by 0x80DED72: CRYPTO_malloc (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x816426E: sk_new (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E00FB: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E0A8A: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x8147926: RSA_new_method (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x5CADD56: _libssh2_rsa_new (openssl.c:69)
==3020==    by 0x5C99785: hostkey_method_ssh_rsa_init (hostkey.c:95)
==3020==    by 0x5C99E4E: diffie_hellman_sha1 (kex.c:267)
==3020==    by 0x5C9B2A6: kex_method_diffie_hellman_group14_sha1_key_exchange (kex.c:799)
==3020==    by 0x5C9BD89: _libssh2_kex_exchange (kex.c:1770)
==3020==    by 0x5CA2756: libssh2_session_handshake (session.c:723)
==3020== 
==3020== 32 bytes in 1 blocks are still reachable in loss record 4 of 6
==3020==    at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==3020==    by 0x80DED72: CRYPTO_malloc (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x816428C: sk_new (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E00FB: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E0A8A: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x8147926: RSA_new_method (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x5CADD56: _libssh2_rsa_new (openssl.c:69)
==3020==    by 0x5C99785: hostkey_method_ssh_rsa_init (hostkey.c:95)
==3020==    by 0x5C99E4E: diffie_hellman_sha1 (kex.c:267)
==3020==    by 0x5C9B2A6: kex_method_diffie_hellman_group14_sha1_key_exchange (kex.c:799)
==3020==    by 0x5C9BD89: _libssh2_kex_exchange (kex.c:1770)
==3020==    by 0x5CA2756: libssh2_session_handshake (session.c:723)
==3020== 
==3020== 128 bytes in 1 blocks are still reachable in loss record 5 of 6
==3020==    at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==3020==    by 0x80DED72: CRYPTO_malloc (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x8164961: lh_new (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E0034: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E00C4: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E0A8A: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x8147926: RSA_new_method (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x5CADD56: _libssh2_rsa_new (openssl.c:69)
==3020==    by 0x5C99785: hostkey_method_ssh_rsa_init (hostkey.c:95)
==3020==    by 0x5C99E4E: diffie_hellman_sha1 (kex.c:267)
==3020==    by 0x5C9B2A6: kex_method_diffie_hellman_group14_sha1_key_exchange (kex.c:799)
==3020==    by 0x5C9BD89: _libssh2_kex_exchange (kex.c:1770)
==3020== 
==3020== 176 bytes in 1 blocks are still reachable in loss record 6 of 6
==3020==    at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==3020==    by 0x80DED72: CRYPTO_malloc (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x816493F: lh_new (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E0034: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E00C4: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x80E0A8A: ??? (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x8147926: RSA_new_method (in /lib/x86_64-linux-gnu/libcrypto.so.1.0.0)
==3020==    by 0x5CADD56: _libssh2_rsa_new (openssl.c:69)
==3020==    by 0x5C99785: hostkey_method_ssh_rsa_init (hostkey.c:95)
==3020==    by 0x5C99E4E: diffie_hellman_sha1 (kex.c:267)
==3020==    by 0x5C9B2A6: kex_method_diffie_hellman_group14_sha1_key_exchange (kex.c:799)
==3020==    by 0x5C9BD89: _libssh2_kex_exchange (kex.c:1770)
==3020== 
==3020== LEAK SUMMARY:
==3020==    definitely lost: 0 bytes in 0 blocks
==3020==    indirectly lost: 0 bytes in 0 blocks
==3020==      possibly lost: 0 bytes in 0 blocks
==3020==    still reachable: 416 bytes in 6 blocks
==3020==         suppressed: 0 bytes in 0 blocks
==3020== 
==3020== For counts of detected and suppressed errors, rerun with: -v
==3020== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
_______________________________________________
libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel

Gmane