Recherche avancée

Sur d’autres sites (4944)

• ffmpeg hangs when run in a multi-threaded environemt

  13 mai 2016, par Zaid Amir

  I have a service that needs to transcode large amount of videos with different formats. The service spawns five threads, one for a single video and each thread runs ffmpeg with the following command :

  ffmpeg -i %%FILEPATH%% -vf scale=X:Y -ab 128k -c:a aac -movflags faststart -strict -2 -ar 22050 -r 24 -c:v libx264 -crf 25 -y %%OUTPUT.MP4%%

  where X and Y are the desired dimensions based on the orientation of the original file basically its either 640:trunc(ow*a/2)*2 for landscape or trunc(oh*a/2)*2:640 for portrait.

  This is my ffmpeg info :

  ffmpeg version 2.4.3-1ubuntu1~trusty6 Copyright (c) 2000-2014 the FFmpeg developers
  
    built on Nov 22 2014 17:07:19 with gcc 4.8 (Ubuntu 4.8.2-19ubuntu1)
  
    configuration: --prefix=/usr --extra-version='1ubuntu1~trusty6' --build-suffix=-ffmpeg --toolchain=hardened --extra-cflags= --extra-cxxflags= --libdir=/usr/lib/x86_64-linux-gnu --shlibdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --enable-gpl --enable-shared --disable-stripping --enable-avresample --enable-avisynth --enable-fontconfig --enable-gnutls --enable-ladspa --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libcdio --enable-libflite --enable-libfreetype --enable-libfribidi --enable-libgme --enable-libgsm --enable-libmodplug --enable-libmp3lame --enable-libopenjpeg --enable-libopus --enable-libpulse --enable-librtmp --enable-libschroedinger --enable-libshine --enable-libspeex --enable-libssh --enable-libtheora --enable-libtwolame --enable-libvorbis --enable-libvpx --enable-libwavpack --enable-libwebp --enable-opengl --enable-x11grab --enable-libxvid --enable-libx265 --enable-libdc1394 --enable-libiec61883 --enable-libzvbi --enable-libzmq --enable-frei0r --enable-libx264 --enable-libsoxr --enable-openal --enable-libopencv
  
    libavutil      54.  7.100 / 54.  7.100
  
    libavcodec     56.  1.100 / 56.  1.100
  
    libavformat    56.  4.101 / 56.  4.101
  
    libavdevice    56.  0.100 / 56.  0.100
  
    libavfilter     5.  1.100 /  5.  1.100
  
    libavresample   2.  1.  0 /  2.  1.  0
  
    libswscale      3.  0.100 /  3.  0.100
  
    libswresample   1.  1.100 /  1.  1.100
  
    libpostproc    53.  0.100 / 53.  0.100
  
  Hyper fast Audio and Video encoder
  
  usage: ffmpeg [options] [[infile options] -i infile]... {[outfile options] outfile}...

  The service is written in Java and running on Ubuntu Server 14.04 and the machine is an 64-bit octa-core server.

  This is the code block that executes ffmpeg :

  try 
  
  {
  
      ProcessBuilder procBuilder = null;
  
  
  
  
  
      String sArgs = String.format("ffmpeg -i %s -vf scale=%s:%s -ab 128k -c:a aac -movflags faststart -strict -2 -ar 22050 -r 24 -c:v libx264 -crf 25 -y %s",
  
                  originalPath,
  
                  outWidth,
  
                  outHeight,
  
                  targetPath
  
          );
  
  
  
      }
  
  
  
      String[] arrArgs = sArgs.split("\\s+");
  
      procBuilder = new ProcessBuilder(Arrays.asList(arrArgs));
  
      procBuilder.redirectErrorStream(true);
  
     procBuilder.redirectOutput();
  
  
  
  
  
      Process process = procBuilder.start();
  
  
  
      try (BufferedReader reader = new BufferedReader(new InputStreamReader(process.getInputStream()))) 
  
      {
  
          String line = null;
  
          while ((line = reader.readLine()) != null) 
  
          {
  
              System.out.println(line);
  
          }
  
          errorCode = process.waitFor();
  
      }
  
  }
  
  catch(Throwable ex)
  
  {
  
  }

  I am currently spawning five threads, and each thread runs a single instance of ffmpeg targeting a single video file. This works fine most of the time, but every once in a while threads start to hang. I noticed from top that ffmpeg hangs indefinitely on some files with one of the threads using 100% of the core CPU and no progress is made. It happened with different file types as I noticed this on mkv, avi, wmv and mp4 files.

  I am not sure what is causing ffmpeg to hang, it does not happen right at the beginning of the transcoding process, ffmpeg starts converting the file fine but somewhere in the middle it gets stuck.

  Now this is not a problem with the files as when I try the same command on the same file manually it runs fine. And it only seems to happen when there are multiple instances of ffmpeg running at the same time as I now changed my service to only run a single thread and it has been running for almost a month with no issues.

  Is there an option that I need to use to allow multiple instances of ffmpeg to run at the same time ? Is it something in the command line that I currently use that causes this ?

• ffmpeg hangs when run in a multi-threaded environment

  3 février 2018, par Zaid Amir

  I have a service that needs to transcode large amount of videos with different formats. The service spawns five threads, one for a single video and each thread runs ffmpeg with the following command :

  ffmpeg -i %%FILEPATH%% -vf scale=X:Y -ab 128k -c:a aac -movflags faststart -strict -2 -ar 22050 -r 24 -c:v libx264 -crf 25 -y %%OUTPUT.MP4%%

  where X and Y are the desired dimensions based on the orientation of the original file basically its either 640:trunc(ow*a/2)*2 for landscape or trunc(oh*a/2)*2:640 for portrait.

  This is my ffmpeg info :

  ffmpeg version 2.4.3-1ubuntu1~trusty6 Copyright (c) 2000-2014 the FFmpeg developers
  
    built on Nov 22 2014 17:07:19 with gcc 4.8 (Ubuntu 4.8.2-19ubuntu1)
  
    configuration: --prefix=/usr --extra-version='1ubuntu1~trusty6' --build-suffix=-ffmpeg --toolchain=hardened --extra-cflags= --extra-cxxflags= --libdir=/usr/lib/x86_64-linux-gnu --shlibdir=/usr/lib/x86_64-linux-gnu --incdir=/usr/include/x86_64-linux-gnu --enable-gpl --enable-shared --disable-stripping --enable-avresample --enable-avisynth --enable-fontconfig --enable-gnutls --enable-ladspa --enable-libass --enable-libbluray --enable-libbs2b --enable-libcaca --enable-libcdio --enable-libflite --enable-libfreetype --enable-libfribidi --enable-libgme --enable-libgsm --enable-libmodplug --enable-libmp3lame --enable-libopenjpeg --enable-libopus --enable-libpulse --enable-librtmp --enable-libschroedinger --enable-libshine --enable-libspeex --enable-libssh --enable-libtheora --enable-libtwolame --enable-libvorbis --enable-libvpx --enable-libwavpack --enable-libwebp --enable-opengl --enable-x11grab --enable-libxvid --enable-libx265 --enable-libdc1394 --enable-libiec61883 --enable-libzvbi --enable-libzmq --enable-frei0r --enable-libx264 --enable-libsoxr --enable-openal --enable-libopencv
  
    libavutil      54.  7.100 / 54.  7.100
  
    libavcodec     56.  1.100 / 56.  1.100
  
    libavformat    56.  4.101 / 56.  4.101
  
    libavdevice    56.  0.100 / 56.  0.100
  
    libavfilter     5.  1.100 /  5.  1.100
  
    libavresample   2.  1.  0 /  2.  1.  0
  
    libswscale      3.  0.100 /  3.  0.100
  
    libswresample   1.  1.100 /  1.  1.100
  
    libpostproc    53.  0.100 / 53.  0.100
  
  Hyper fast Audio and Video encoder
  
  usage: ffmpeg [options] [[infile options] -i infile]... {[outfile options] outfile}...

  The service is written in Java and running on Ubuntu Server 14.04 and the machine is an 64-bit octa-core server.

  This is the code block that executes ffmpeg :

  try 
  
  {
  
      ProcessBuilder procBuilder = null;
  
  
  
  
  
      String sArgs = String.format("ffmpeg -i %s -vf scale=%s:%s -ab 128k -c:a aac -movflags faststart -strict -2 -ar 22050 -r 24 -c:v libx264 -crf 25 -y %s",
  
                  originalPath,
  
                  outWidth,
  
                  outHeight,
  
                  targetPath
  
          );
  
  
  
      }
  
  
  
      String[] arrArgs = sArgs.split("\\s+");
  
      procBuilder = new ProcessBuilder(Arrays.asList(arrArgs));
  
      procBuilder.redirectErrorStream(true);
  
     procBuilder.redirectOutput();
  
  
  
  
  
      Process process = procBuilder.start();
  
  
  
      try (BufferedReader reader = new BufferedReader(new InputStreamReader(process.getInputStream()))) 
  
      {
  
          String line = null;
  
          while ((line = reader.readLine()) != null) 
  
          {
  
              System.out.println(line);
  
          }
  
          errorCode = process.waitFor();
  
      }
  
  }
  
  catch(Throwable ex)
  
  {
  
  }

  I am currently spawning five threads, and each thread runs a single instance of ffmpeg targeting a single video file. This works fine most of the time, but every once in a while threads start to hang. I noticed from top that ffmpeg hangs indefinitely on some files with one of the threads using 100% of the core CPU and no progress is made. It happened with different file types as I noticed this on mkv, avi, wmv and mp4 files.

  I am not sure what is causing ffmpeg to hang, it does not happen right at the beginning of the transcoding process, ffmpeg starts converting the file fine but somewhere in the middle it gets stuck.

  Now this is not a problem with the files as when I try the same command on the same file manually it runs fine. And it only seems to happen when there are multiple instances of ffmpeg running at the same time as I now changed my service to only run a single thread and it has been running for almost a month with no issues.

  Is there an option that I need to use to allow multiple instances of ffmpeg to run at the same time ? Is it something in the command line that I currently use that causes this ?

• Problems with slowing down when grid synthesis operation request via ffmpeg is received simultaneously

  7 novembre 2022, par bongcheol Kim
  

  1. Crop/scaling/volume leveling encoding of videos uploaded by users in S3 with predetermined settings

  


  2. Synthesize in grid form with other pre-prepared videos.

  


  


  Perform the above tasks through ffmpeg.
Encoding slows down when multiple requests occur simultaneously.
(Maybe it's because we shared the server's resources.)

  


  To process task requests from 1 and 2 at the same time, but to process them
What can I do ?

  


  Is there a service on AWS that can solve my problems ?

  


  I thought about creating and processing ec2 individually for each request, but is this possible ?

  


  

  1. Invoke API

  


  2. Create EC2

  


  3. Proceed with one encoding operation on the corresponding EC2

  


  4. Delete EC2 after uploading s3