Advanced search

On other websites (6448)

• Continously running a PHP script waiting for videos to trancode

  12 May 2017, by Ian

  I’m making a transcoding server which uses FFMPEG to convert videos to flv. After user uploads a video it’s queued for processing in amazon Simple Queue Service. System is linux ubuntu.

  Instead of running CRON each 1min I wonder if it would be possible to continously run several PHP scripts (dowload queued files, process downloaded etc). Each of them would have its own queue which would be read every 10s or so looking for new tasks.

  My question is:

  How to detect if the script is already running? I’d run CRON each 1min and if one of the programs would not be running I’d load it again. How stuff like that is done on linux? PID files?

  thanks for help,
  ian

• Why does fluent-ffmpeg only work when it throws the error Output stream closed

  29 March 2024, by volume one

  I am using fluent-ffmpeg to process a video file (and then upload that to Amazon S3). The code is very straightforward but it only works if:

  


  

  • pipe option {end: true} is set in .output()

  


  • which has a side-effect that causes the following console log output

  


  


  


  Processing: 19.261847354642416% done Processing:
32.365144874807335% done Processing: 48.80978326261429% done Processing: 78.35771917058617%
Processing: 91.49377493455148% done Processing:
99.91264359125745% done An error occurred: Output stream closed

  


  


  Despite that error, it seems the file is generated correctly and it gets uploaded to Amazon S3 fine.

  


  This is the fluent-ffmpeg code:

  


  import {PassThrough} from 'node:stream';
import FFMpeg from 'fluent-ffmpeg';

let PassThroughStream = new PassThrough();

             FFMpeg('/testvideo.mp4')
                    .videoCodec('libx264')
                    .audioCodec('libmp3lame')
                    .size(`640x480`)
                    // Stream output requires manually specifying output formats
                    .format('mp4')
                    .outputOptions('-movflags dash')
                    .on('progress', function (progress) {
                        console.log('Processing: ' + progress.percent + '% done');
                    })
                    .on('error', function (err) {
                        console.log('An error occurred: ' + err.message);
                    })
                    .on('end', function () {
                        console.log('FFMpeg Processing finished!');
                    })
                    .output(PassThroughStream, {end: true})
                    .run();

   // Now upload to S3
    try {
          await s3Upload({
              AWSS3Client: 'mys3client',
              Bucket: 'publicbucket,
              ACL: "public-read",
              ContentType: 'video/mp4',
              Key: 'whoever/whatever.mp4',
              Body: PassThroughStream
           });
    } catch (error) {
                    console.log(`s3Upload error`, error)
                }


  


  If I set the pipe output() option to {end: false} then there is no error from fluent-ffmpeg and I get "Processing: 100% done FFMpeg Processing finished!" as the final console log.

  


  BUT the problem is that the s3Upload() does not do anything. There are no errors. Just no activity.

  


  I feel very uncomfortable letting fluent-ffmpeg end in an error even if the code itself does the job intended. It will also cause testing to fail. What could be the issue?

  


  The command line code is: ffmpeg -i https:/xxxbucket.s3.amazonaws.com/14555/file-example.mp4 -acodec libmp3lame -vcodec libx264 -filter:v scale=w=trunc(oh*a/2)*2:h=480 -f mp4 -movflags dash pipe:1

  


• stream usb webcam via http using vlc without transcoding using mjpeg signal from webcam

  10 November 2020, by Peter Nußreiner

  I wanna stream my (Microsoft live) USB webcam using vlc from a raspberry.
While it works perfectely fine with transcoding my raspberry does not have enough power.

  


  So I checked the output formats of my webcam using:

  


  ffmpeg -f v4l2 -list_formats all -i /dev/video0


  


  wich gives me the following output:

  


  [video4linux2,v4l2 @ 0x559d5aee36c0] Raw       :     yuyv422 :           YUYV 4:2:2 : 640x480 1280x720 960x544 800x448 640x360 424x240 352x288 320x240 800x600 176x144 160x120 1280x800
[video4linux2,v4l2 @ 0x559d5aee36c0] Compressed:       mjpeg :          Motion-JPEG : 640x480 1280x720 960x544 800x448 640x360 800x600 416x240 352x288 176x144 320x240 160x120
/dev/video0: Immediate exit requested


  


  when running my streaming command:

  


  cvlc v4l2:///dev/video0:chroma=h264:width=1280:height=720 --sout '#standard{access=http,mux=ts,dst=192.168.178.27:8080,name=stream,mime=video/ts}' -vvv


  


  I get the following error:

  


  mux_ts mux warning: rejecting stream with unsupported codec YUY2
main mux error: cannot add this stream
main decoder error: cannot create packetizer output (YUY2)


  


  along with many other lines of information.
As far as I can interpret the error, vlc wants to use the YUY2 RAW signal of my webcam.
Is there any way I can force vlc to use the mjpg output of my camera?

  


  I've already tried to use 'chroma=mjpg' without success.
Transcoding is not an alternative as my raspi has by far not enough power.

  


  I've also read about using ffmpeg for streaming, maybe someone has an alternative solution since I'm very new to the whole topic of websites and stuff

  


  It will (or should) be used as a live webcam for a windsurfing spot.
Therefore I need a fluent video with at least 10fps to judge the situation. A delay of a minute or so is no problem.
The basic idea was to stream the video to a webserver and embed the image in a website. I've got good internet connection at the spot.

  


  But for now I want to try it locally on my pc.