Memory leak when I put for recording (smart recording)

[junshengy]

@s.Jagannath

cache_que0 with this, how can I get the cache video?

@s.Jagannath

How do I get the queue which is present after the Tee so that,

I don’t understand what you mean, cache_que0 is what you need. If you want to get the input and output data of the element, please add a probe function on the src/sink pad

You can draw the entire pipeline as a picture in the following way

DeepStream SDK FAQ

5. [DS5.0GA_App] Generate GStreamer Pipeline Graph Use one of below method according to your application type to generate the GST pipeline graph. 4.1 deepstream-app run "export GST_DEBUG_DUMP_DOT_DIR=/tmp/" before deepstream-app command, e.g. $ sudo apt-get install graphviz $ export GST_DEBUG_DUMP_DOT_DIR=/tmp/ $ deepstream-app -c deepstream_app_config_yoloV2.txt $ cd /tmp/ $ dot -Tpng 0.03.47.898178403-ds-app-playing.dot >~/0.03.47.898178403-ds-app-playing.png // png file includes th…

If you want to record overlapped videos, I have also provided a solution before

How to record overlapped videos

This requires some special operations,need to use multiple nvurisrcbin elements. Add multiple nvurisrcbin elements to your pipeline, and then connect the multiple nvurisrcbin elements to the same source stream. When you want to record overlapped videos, send the event to the second/third nvurisrcbin in turn to avoid first nvurisrbin’s buffer clearing.

[junshengy]

As I replied before, if you want to trigger start-sr continuously, cache_que0 will not be able to get the data you need.

Can you share your goal? Or try the overlap recording solution above.

[s.Jagannath]

our goal is that sometimes we need video and sometimes image based on our usecase . SO, How can I get an image from the smart record cache queue?

[junshengy]

If smart-recording can record the video you need, you can consider extracting pictures at a specified time according to pts from the recorded video.

[s.Jagannath]

I found below code at /opt/nvidia/deepstream/deepstream-7.0/sources/apps/apps-common/src/deepstream_source_bin.c and how can I use this code in python?

void
destroy_smart_record_bin (gpointer data)
{
  unsigned int i = 0;
  NvDsSrcBin *src_bin;
  NvDsSrcParentBin *pbin = (NvDsSrcParentBin *) data;

  g_return_if_fail (pbin);

  for (i = 0; i < pbin->num_bins; i++) {

    src_bin = &pbin->sub_bins[i];
    if (src_bin && src_bin->recordCtx)
      NvDsSRDestroy ((NvDsSRContext *) src_bin->recordCtx);
  }
  pbin->num_bins = 0;
}

static gpointer
smart_record_callback (NvDsSRRecordingInfo * info, gpointer userData)
{
  static GMutex mutex;
  FILE *logfile = NULL;
  g_return_val_if_fail (info, NULL);

  g_mutex_lock (&mutex);
  logfile = fopen ("smart_record.log", "a");
  if (logfile) {
    fprintf (logfile, "%d:%d:%d:%ldms:%s:%s\n",
        info->sessionId, info->width, info->height, info->duration,
        info->dirpath, info->filename);
    fclose (logfile);
  } else {
    g_print ("Error in opeing smart record log file\n");
  }
  g_mutex_unlock (&mutex);

  return NULL;
}

[junshengy]

@s.Jagannath

NvDsSRDestroy ((NvDsSRContext *) src_bin->recordCtx);

nvurisrcbin is responsible for calling NvDsSRDestroy. In addition, the sr-done signal is the same as smart_record_callback. These codes are useless in Python.

[Fiona.Chen]

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks