Recherche avancée

Sur d’autres sites (6369)

• how is the 'response' parameter calculated for rtsp authentication ?

  11 septembre 2015, par zcaudate

  So I’ve captured a couple of packets from ffmpeg to a ip camera with rtsp, which I’ve attached at the end of the question :

  In the second request/response pair, the protocol sees that it needs authentication and returns a nonce="be129bc71fa3ddfbff9989ed6f748268" to the user

  In the third request/response pair, there is a response="da0882d360a56deef2994d8ba0de8e89" being sent to the server and I’m not sure how that is generated. The user/password pair is admin with a blank password.

  Any ideas of where to look in the RFCs would be great

  ---

  OPTIONS rtsp://192.168.1.245:88/videoMain RTSP/1.0
  
  CSeq: 1
  
  User-Agent: Lavf56.4.101
  
  
  
  RTSP/1.0 200 OK
  
  CSeq: 1
  
  Date: Sun, Sep 21 2014 01:33:07 GMT
  
  Public: OPTIONS, DESCRIBE, SETUP, TEARDOWN, PLAY, PAUSE, GET_PARAMETER, SET_PARAMETER
  
  
  
  ----------
  
  
  
  DESCRIBE rtsp://192.168.1.245:88/videoMain RTSP/1.0
  
  Accept: application/sdp
  
  CSeq: 2
  
  User-Agent: Lavf56.4.101
  
  
  
  RTSP/1.0 401 Unauthorized
  
  CSeq: 2
  
  Date: Sun, Sep 21 2014 01:33:07 GMT
  
  WWW-Authenticate: Digest realm="LIVE555 Streaming Media", nonce="be129bc71fa3ddfbff9989ed6f748268"
  
  
  
  ----------
  
  
  
  DESCRIBE rtsp://192.168.1.245:88/videoMain RTSP/1.0
  
  Accept: application/sdp
  
  CSeq: 3
  
  User-Agent: Lavf56.4.101
  
  Authorization: Digest username="admin",realm="LIVE555 Streaming Media",nonce="be129bc71fa3ddfbff9989ed6f748268",uri="rtsp://192.168.1.245:88/videoMain",response="da0882d360a56deef2994d8ba0de8e89"
  
  
  
  
  
  RTSP/1.0 200 OK
  
  CSeq: 3
  
  Date: Sun, Sep 21 2014 01:33:07 GMT
  
  Content-Base: rtsp://192.168.1.245:65534/videoMain/
  
  Content-Type: application/sdp
  
  Content-Length: 500
  
  
  
  v=0
  
  o=- 1411259869717503 1 IN IP4 192.168.1.245
  
  s=IP Camera Video
  
  i=videoMain
  
  t=0 0
  
  a=tool:LIVE555 Streaming Media v2013.01.25
  
  a=type:broadcast
  
  a=control:*
  
  a=range:npt=0-
  
  a=x-qt-text-nam:IP Camera Video
  
  a=x-qt-text-inf:videoMain
  
  m=video 0 RTP/AVP 96
  
  c=IN IP4 0.0.0.0
  
  b=AS:96
  
  a=rtpmap:96 H264/90000
  
  a=fmtp:96 packetization-mode=1;profile-level-id=42001F;sprop-parameter-sets=Z0IAH5WoFAFuQA==,aM48gA==
  
  a=control:track1
  
  m=audio 0 RTP/AVP 0
  
  c=IN IP4 0.0.0.0
  
  b=AS:64
  
  a=control:track2

• FFMpeg/FFServer P2P streaming between embedded device and smartphone

  25 avril 2016, par Bogdan Emil Mariesan

  I’ve been playing for the last couple of days with FFMpeg and FFServer as I am considering them as candidates for livestreaming between an embedded device(with an integrated HD camera) and various clients(smartphones).

  I’ve managed to achieve the stream using with the following config for FFServer :

  HTTPPort 1234
  
  RTSPPort 1235
  
  MaxHTTPConnections 2000
  
  MaxClients 1000
  
  MaxBandwidth 512000             # Maximum bandwidth per client
  
                                 # set this high enough to exceed stream bitrate
  
  CustomLog -
  
  NoDaemon                       # Remove this if you want FFserver to daemonize after start
  
  <feed>               # This is the input feed where FFmpeg will send
  
    File /tmp/feed.ffm            # video stream.
  
    FileMaxSize 512K
  
  </feed>
  
  <stream>                      # Output stream URL definition
  
     Feed feed.ffm              # Feed from which to receive video
  
     Format rtp
  
     # Video settings
  
     VideoCodec libvpx
  
     VideoSize 720x576           # Video resolution
  
     VideoFrameRate 60           # Video FPS
  
     AVOptionVideo flags +global_header  # Parameters passed to encoder
  
                                         # (same as ffmpeg command-line parameters)
  
     AVOptionVideo cpu-used 0
  
     AVOptionVideo qmin 10
  
     AVOptionVideo qmax 42
  
     AVOptionVideo quality good
  
     AVOptionAudio flags +global_header
  
     PreRoll 15
  
     StartSendOnKey
  
     VideoBitRate 400            # Video bitrate
  
     NoAudio
  
  </stream>

  And the following FFMpeg command to send the stream to FFServer :

  ffmpeg -rtbufsize 2100M -f dshow -i video="Integrated Camera" -vcodec libx264 http://127.0.0.1:1234/feed.ffm

  I also have a simple Android client that plays the RTSP stream using the following URL :

  rtsp://mylocalnetworkip:1235/test.h264

  But now I am trying to achieve a P2P connection between the embedded device and a smartphone. This has to be over the internet(not in the LAN) and capable to achieve UDP hole punching (such as Skype does for p2p video-calling).

  1. Is this achievable with ffmpeg alone ?
  2. Can ffmpeg integrate with a Stun/Turn server such as Coturn to bypass symmetric NATs ?

• FFMpeg lib memory leak

  30 mai 2018, par unresolved_external

  I am using lib ffmpeg for video decoding. So in my code I have something like :

  if ((err = av_hwdevice_ctx_create(&m_hw_device_ctx, type,
  
                                              NULL, NULL, 0)) < 0) 
  
      {
  
            printf( "Failed to create specified HW device.\n");
  
            return err;
  
          }
  
  
  
          m_decoder_ctx->hw_device_ctx = av_buffer_ref(m_hw_device_ctx);
  
          if ((ret = avcodec_open2(m_decoder_ctx, decoder, NULL)) < 0)
  
          {
  
              printf("Failed to open codec for video stream\n");
  
              return -1;                
  
          }
  
          if (0 > avcodec_send_packet(m_decoder_ctx, packet)) 
  
          {                
  
              return -1;
  
          }
  
      AVFrame *frame = nullptr;
  
      if (!(frame = av_frame_alloc())) 
  
      {
  
           printf( "Can not alloc frame\n");
  
           return -1;
  
      }
  
  
  
  
  
  
  
      int ret = avcodec_receive_frame(avctx, frame);
  
      if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) 
  
      {
  
          av_frame_free(&frame);
  
          return 0;
  
      } 
  
      else if (ret < 0) 
  
      {
  
          printf("Error while decoding\n");
  
          av_frame_free(&frame);
  
      }
  
  //here is some application specific manipulation(ommitted for redability)
  
    av_frame_free(&frame);

  And then I free allocated memory like this :

  av_buffer_unref(&m_decoder_ctx->hw_device_ctx);
  
              av_buffer_unref(&m_hw_device_ctx);
  
              m_hw_device_ctx = nullptr;

  But Valgrind constantly reports an error :

  ==1115== 524,288 bytes in 1 blocks are definitely lost in loss record 2,147 of 2,174
  
  ==1115==    at 0x76A67920: ??? (in /usr/lib/x86_64-linux-gnu/libdrm_intel.so.1.0.0)
  
  ==1115==    by 0x76727A07: ??? (in /usr/lib/x86_64-linux-gnu/dri/i965_drv_video.so)
  
  ==1115==    by 0x76727B2A: ??? (in /usr/lib/x86_64-linux-gnu/dri/i965_drv_video.so)
  
  ==1115==    by 0x766FC2F5: ??? (in /usr/lib/x86_64-linux-gnu/dri/i965_drv_video.so)
  
  ==1115==    by 0x7670459C: __vaDriverInit_0_39 (in /usr/lib/x86_64-linux-gnu/dri/i965_drv_video.so)
  
  ==1115==    by 0x144A8AE5: ??? (in /usr/lib/x86_64-linux-gnu/libva.so.1.3900.0)
  
  ==1115==    by 0x144A9906: vaInitialize (in /usr/lib/x86_64-linux-gnu/libva.so.1.3900.0)
  
  ==1115==    by 0xD03A55E: ??? (in /usr/local/lib/libavutil.so.56.18.102)
  
  ==1115==    by 0xD03839C: av_hwdevice_ctx_create (in /usr/local/lib/libavutil.so.56.18.102)

  I’ve been trying to troubleshoot it without any success. Also checked a couple of examples - looks like everyone works with that function the same as I do. (e.g. -https://github.com/FFmpeg/FFmpeg/blob/master/doc/examples/vaapi_encode.c)
  Compiler flag, which I am using are :

  -O0 -g3 -pthread -fpermissive -lm

  Any ideas - what is wrong with buffer deallocation ?