Rtsp live stream gets delayed overtime

[edje11]

I'm using @kevinGodell his nodes for my cctv cameras now for a couple of years with great joy.
Bought myself an Reolink doorbel last week and integrated (rtsp stream) this also with Kevin his nodes in my dashboard and Bart his Onvif node to capture bell presses. Nice.
Works all great, however during the day the live stream gets delay with 20sec or something like that, delay will increase overtime.
20 sec delay is to much to see "live" who is in front of the door.
Restarting the ffmpeg node with the restart command will solve this issue and the live feed is back at his "normal" 5 sec delay.
Is there a better / neater way to get the everything running on time?

[{"id":"5654e6d653ad2b72","type":"inject","z":"44908087.da8e9","name":"stop","props":[{"p":"action","v":"{\"command\":\"stop\"}","vt":"json"}],"repeat":"","crontab":"","once":false,"onceDelay":0.1,"topic":"","x":390,"y":4680,"wires":[["7a8c3f95b486b8f9"]]},{"id":"3db9c9f03c8f6cae","type":"inject","z":"44908087.da8e9","name":"start","props":[{"p":"action","v":"{\"command\":\"start\"}","vt":"json"}],"repeat":"","crontab":"","once":false,"onceDelay":0.1,"topic":"","x":390,"y":4560,"wires":[["7a8c3f95b486b8f9"]]},{"id":"30bbbddd32d43bde","type":"debug","z":"44908087.da8e9","name":"stderr","active":false,"tosidebar":true,"console":false,"tostatus":false,"complete":"true","targetType":"full","statusVal":"","statusType":"auto","x":820,"y":4780,"wires":[]},{"id":"4cc7afbbcc802243","type":"mp4frag","z":"44908087.da8e9","name":"","outputs":2,"basePath":"ui_mp4frag_01","serveHttp":"true","serveIo":"true","hlsPlaylistSize":"6","hlsPlaylistExtra":"2","autoStart":"false","preBuffer":1,"timeLimit":10000,"repeated":"false","statusData":"playlist","x":890,"y":4620,"wires":[["91780c25f9e1edae"],[]]},{"id":"91780c25f9e1edae","type":"ui_mp4frag","z":"44908087.da8e9","name":"Camera Deur","group":"7ea407c1b15941c7","order":1,"width":"21","height":"11","readyPoster":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAyAAAAJYCAMAAACtqHJCAAAAGXRFWHRTb2Z0d2FyZQBBZG9iZSBJbWFnZVJlYWR5ccllPAAAA2hpVFh0WE1MOmNvbS5hZG9iZS54bXAAAAAAADw/eHBhY2tldCBiZWdpbj0i77u/IiBpZD0iVzVNME1wQ2VoaUh6cmVTek5UY3prYzlkIj8+IDx4OnhtcG1ldGEgeG1sbnM6eD0iYWRvYmU6bnM6bWV0YS8iIHg6eG1wdGs9IkFkb2JlIFhNUCBDb3JlIDUuMy1jMDExIDY2LjE0NTY2MSwgMjAxMi8wMi8wNi0xNDo1NjoyNyAgICAgICAgIj4gPHJkZjpSREYgeG1sbnM6cmRmPSJodHRwOi8vd3d3LnczLm9yZy8xOTk5LzAyLzIyLXJkZi1zeW50YXgtbnMjIj4gPHJkZjpEZXNjcmlwdGlvbiByZGY6YWJvdXQ9IiIgeG1sbnM6eG1wTU09Imh0dHA6Ly9ucy5hZG9iZS5jb20veGFwLzEuMC9tbS8iIHhtbG5zOnN0UmVmPSJodHRwOi8vbnMuYWRvYmUuY29tL3hhcC8xLjAvc1R5cGUvUmVzb3VyY2VSZWYjIiB4bWxuczp4bXA9Imh0dHA6Ly9ucy5hZG9iZS5jb20veGFwLzEuMC8iIHhtcE1NOk9yaWdpbmFsRG9jdW1lbnRJRD0ieG1wLmRpZDowMzgwMTE3NDA3MjA2ODExODIyQUIzREQ3RTA0MDAyNCIgeG1wTU06RG9jdW1lbnRJRD0ieG1wLmRpZDpGRkU5MkM4RDJCNTkxMUVCQjJCOTg0MEQ0QThCQ0YyQyIgeG1wTU06SW5zdGFuY2VJRD0ieG1wLmlpZDpGRkU5MkM4QzJCNTkxMUVCQjJCOTg0MEQ0QThCQ0YyQyIgeG1wOkNyZWF0b3JUb29sPSJBZG9iZSBQaG90b3Nob3AgQ1M2IChNYWNpbnRvc2gpIj4gPHhtcE1NOkRlcml2ZWRGcm9tIHN0UmVmOmluc3RhbmNlSUQ9InhtcC5paWQ6Rjc3RjExNzQwNzIwNjgxMTgyMkFDQTkzQjdGQkY1MzAiIHN0UmVmOmRvY3VtZW50SUQ9InhtcC5kaWQ6MDM4MDExNzQwNzIwNjgxMTgyMkFCM0REN0UwNDAwMjQiLz4gPC9yZGY6RGVzY3JpcHRpb24+IDwvcmRmOlJERj4gPC94OnhtcG1ldGE+IDw/eHBhY2tldCBlbmQ9InIiPz7USCkmAAAAM1BMVEUAplH///9Cs3HR6dqc0rDp9O1rwIy43sUmrWJ9xpn0+vbF5NBYun+q2LuNzKTd7+P///8HAxLBAAAAEXRSTlP/////////////////////ACWtmWIAAA6fSURBVHja7d1rY6JIGoBRI6gIXub//9qZ7o4KRdVbZezuJLvnfNpJektujyAg2fwDFG0sAhAICAQEAgIBgYBAQCAgEBAIIBAQCAgEBAICAYGAQEAgIBAQCCAQEAgIBAQCAgGBgEBAICAQQCAgEBAICAQEAgIBgYBAQCAgEEAgIBAQCAgEBAICAYGAQEAggEBAICAQEAgIBAQCAgGBgEBAIIBAQCAgEBAICAQEAgIBgYBAAIGAQEAgIBAQCAgEBAICAYGAQACBgEBAICAQEAgIBAQCAgGBgEAsAhAICAQEAgIBgYBAQCAgEBAIIBAQCAgEBAICAYGAQEAgIBAQCCAQEAgIBAQCAgGBgEBAICAQQCAgEBAICAQEAgIBgYBAQCAgEEAgIBAQCAgEBAICAYGAQEAggEBAICAQEAgIBAQCAgGBgEBAIIBAQCAgEBAICAQEAgIBgYBAAIGAQEAgIBAQCAgEBAICAYGAQACBgEBAICAQEAgIBAQCAgGBgEAsAhAICAQEAgIBgYBAQCAgEBAIIBAQCAgEBAICAYGAQEAgIBAQCCAQEAgIBAQCAgGBgEBAICAQQCAgEBAICAQEAgIBgYBAQCAgEEAgIBAQCAgEBAICAYGAQEAggEBAICAQEAgIBAQCAgGBgEBAIIBAQCAgEBAICAQEAgIBgYBAAIGAQEAgIBAQCAgEBAICAYGAQIAkEAAAAAAAAAAAAAAAAAAAAAAAgD/v0u/efjj0228/L33/c1be+t56/d92+rWi33aZnw2b7fv/6n7+/PZfN12/n/2/0t++b0H3PMbZT6dsIskI1/64+lUXzsQP558/OdxeaL5Rv/9su57WTLPrMdcjJb9cT+TwPtO7IZrR5ctXlmNxwsprp7IwCN2229mPjr9+Mm4qgfxY88e2FTt0yc+P9UB+TMKlKZDLvL35LPyYh7vpMURmWrt9bcz7ZF6TkLfFibwNcqnN6Nt13xpIacLKa6eyMAjd1vZp9e7SNQTy37tfy4odrqtfXFoC+e+Vh4ZAzvONYtn9Y7aG2RRt422wMOb7OLvVdrgtBbLL/P9rL18LpDRhwdqJFwahaf2Wfrgv3IZA7oVEK/aQ+c22KZDkpbtwHma7wmm1BRxnW0l+Wi+VMTO/WEzVaiILO5C4gVogpQkL1k68MIg/bq7fS673n7QEcksrWLH78pbfuN3EgSze00+LbXNK33mv5Vcat/GYm+K87POBFHYg8cvXAilNWLB24oVB6Lj+vDfeF21TILvq5j1lf3Vq3G6GWiBDptf7z4Yk+3MwJ3085k9d+26utAMpvfy5JZDihAVrJ14YNJ3G6lbbxrYxkPe1VF6xw/xD43AonRuKNtswkFNuU7omm9Awf7MvvNJYGTOdl+1hcbyYTmRpBxK/fCWQ0oSFaydcGISG1ee9/eP4Nh/Ir386nJefQsrb8HG+L3gc1F1L282v8ftx9u/CQI6LDWJaHjqeV6fmVsNdduudWnbM9byc5zuJZCKLO5D45eOjyeKEhWsnXBjEdunu91LaLrdJS/3iv8sr9pyc8h3TV8yu4sfR/lDZavrlCdPlO+11ORFTblq364/p2TEz8/KWOXDrFgdju00lkOTlK4GUJixcO+HCINal756zpRcHsjy0La/Y9ETZtfAhJB1/9u/CreZ2Tn95ZicJ/zpPIB1uKp0r6tKzRYV5mdaj7uvns7MvXwmkNGHx2okWBrFz+nlvKh36p6vgvgVv4xWbJjgVjoDT8Q+PfxduNb8mY0wuDi/n67TYQtLh1qfy8mMW5+W6HrW8A4lfvhJIacLitRMtDJ48zzt7e6kE0uU++JYD2a43/KZALpWt5n0bPS5LX56e6xfHGKXzTX1lzOK87FajBjuQ4ssfGgIpTVi8dqKFQWyfXhLPXFD70oG8vx9Op2QrH+ejTdmLccVASmO2BxLsQF4JpDhhlbUTLAxi2+QD3Ha2//0WgRxvqzspfZofDY2LI6NqIKUxmwOJdiCvBFKcsMraCRYGFePynN9+9p/fIpD+9q92y9LnW/1pOXI1kNKYzYFEO5BXAilOWGXtBAuDiuvyI1s/Wz/fIpDD7f1wWpY+zK4V9MvrANVr3qUxWwPZRjuQuM84kOKEVdZOsDCoSLbXc/HM/hcN5Hrr+5y8zvWxDU3Lj7Sl00j76piNgRzCN+kXAilOWGXtBAuDxtNYl9VZ3u8RyHhb831yZqd/HHePy7fbdLhucXk8GrMLvmj0GHUb303+QiDFCasFUl4YVByXa3M3W/DfIZDt/Tf7ZLN8nOk5JleOk+GOb/nfZ8ZsC+SQBtd2HWRfDaQ8YbVAyguDitNy8yhf/PuagRzvH3CH9C7I3W3GzskngsVw2/NbOkHlMZsCqexA4h1Yy21nuZmtBFJeGNQszvOeyjf1fVYg8ZX0/rExjsk/Ot82vGty5Xgb3G4ej9kUSGUHUtqB1W76jGe2FkhxYVCzmzdxnNfykUDWXzb8aCBt92IdVnd1j+nB46V0SjT/5eFozJZAdrXvsxZ2YKdNvBzjma0FUlwY1Ezzld6X73r9y4E03s3bPbauQ3qP0bjYYvsokPHSNGZLIG+177OGO7AwkGBma4EUFwY15/mSPMyX32cG0vp9kPFxW+vqbO1U2GBzW+h5aBnzmUDOTwUSfLf/0DKz1UCmtnpZWZxl7HLfEvy7gTz1jcJhNk3HwgXx1UWJ7CuNp4YxnwlkfCaQy6YhkGhmq4GUFgY1+/k73pg7Hv7MQCrfSd/PDglP6Vv3UHhDrzy0IRjzmUCKJ4rCI7wokGhmq4EMTXs3NqU34Z/LfQi+N/E5gVSuns3vmxhWX0S9PvWAg64+5lOBdM8cYnUNgUQzWw2ktDCoGldfztxtvkogtedinecHGqvHJPTJnih7nvXYvWXPiObGfCqQ0oM9831O9UCima0HUlgYVF0f39+8ZFfW370Osuij9mTFbj5Wl57GOmWfb1D8zuuhOuZzgZyfCaT0dK3Gma0HUlgYVB0eZzb67CnHzwqk4dm843zs8+oUzZh7Qs5quO3yY3Uw5nOBjA3XQfbTcqut33aWn9l6IIWFQdXsxrdD8M27vxzIruXp7sNiS+xXn44fT7IdgkCWt/xHY7YFMh6eud29fyvfvdA+sw2B5BcGVbMzhtfFQv3cW02iLTo5bLguZ+S8mrXllePVcIvnSERjtgXSn8KP6enLj43PxQpntiGQ/MKg6vQ4BB+zx7VfN5D8Z6bHIfb9IVx9ONxhfnInGrPl+yA/PgJfo4/p6cufG5+LFc5sQyD5hUHdfZ0sd+LfIJDzco2vR+hyL1YM5FIbsymQS/aZ4OWXb33sTzizDYHkFwbNp7F2tytR128TSP652G8vBRKN2RLI/HnRY0sgrd9JD2dWIH/QbckPl/x9QV83kF1+mzm9Ekg0ZvNjfw7BFbn05feNgYQzK5A/6H7G8Jy/v+fLBjLkN5nZZvl8IOGYTz84rvt9gcQzK5A/6LaPP4YPN/iCgZzeyjc4fjSQcMzmQO5v99vfFkg8swL5C6ex+uvyCOXLB3IsbDPTC4GEY7YHEnxM/2Ag8cwK5A+67bwPyWfcz314dUMgfWGbuT4ZyPyh7eGYhXnZNSytlwOJZzZeOwJ5ze3pisky/tw/f9AQSOG8zuw0Vlsg80t14Zhp3N38y7PLUbvix/QPBhLPbLx2BPKaLn+A8ql/QKclkNUfulhd6W4K5DjfmMIxk3m5lZXbtC/FGwM/GEg8s/HaEchrDvmbUD/zT7A1BbLaF63OrrYE0i8OVsIxb/OyG9YfNEp3kGx/UyDxzMZrRyCv6eOvf/7dP+IZBLK8mzHzl2BWH43DQHJTG4+5nJflA0lKd5D0HwpkNWWVmd1Wz3gL5OOOmS8mbD7rz0C3B7L+Y5TH9H3zmUB+7uIqY3blRZBu96fS/HwskMqExWtHIK9ZLt3hqUCO1UD20VdMXwmkXw2V/rnKpwLpZ2/LpTH39SsRq5voj78lkMqEbes7EIF8XP5rPi2BnDfVQJKPOKULaM8HclgfrKXz8EQg03xai2NO1X1olx4CTY2BdOFyrExY7WmRAnnJNfve3hDIYdMQyHB9a3jix/OBXNcH+ekF7PZApsWSKI457EoHi6tASufrSsdi13A5Vias9rRIgbxkyi7VaiC746YlkM3QRXv+DwcyrgdLL7q0BnKfk+qY26T2cV881XbIf0xP/+HikkVxOVYmLF47AvmNp7H6xkC6fl89bHk8w22cv1mH3yNqDuSUOVpLTx01BTKbk5Yxz4s36aF8Lnq/PClc/IfTbLdaWo61CYvXjkC+vkv/64jg0G+///vJ7eEWvpsHAAAAAAAAAAAAAAAAAAAAAAAAwP+Xf4AigYBAQCAgEBAICAQEAgIBgYBAAIGAQEAgIBAQCAgEBAICAYGAQACBgEBAICAQEAgIBAQCAgGBAAIBgYBAQCAgEBAICAQEAgIBgQACAYGAQEAgIBAQCAgEBAICAQQCAgGBgEBAICAQEAgIBAQCAgEEAgIBgYBAQCAgEBAICAQEAggEBAICAYGAQEAgIBAQCAgEBAIIBAQCAgGBgEBAICAQEAgIBBAICAQEAgIBgYBAQCAgEBAICAQQCAgEBAICAYGAQEAgIBAQCAgEEAgIBAQCAgGBgEBAICAQEAggEBAICAQEAgIBgYBAQCAgEBAIIBAQCAgEBAICAYGAQEAgIBBAICAQEAgIBAQCAgGBgEBAICAQQCAgEBAICAQEAgIBgYBAQCCAQEAgIBAQCAgEBAICAYGAQEAggEBAICAQEAgIBAQCAgGBgEAAgYBAQCAgEBAICAQEAgIBgYBAAIGAQEAgIBAQCAgEBAICAYGAQACBgEBAICAQEAgIBAQCAgGBAAIBgYBAQCAgEBAICAQEAgIBgQACAYGAQEAgIBAQCAgEBAICAQQCAgGBgEBAICAQEAgIBAQCAgEEAgIBgYBAQCAgEBAICAQEAggEBAICAYGAQEAgIBAQCAgEBAIk/gX38R0SGa0LJgAAAABJRU5ErkJggg==","errorPoster":"data:image/png;base64,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","hlsJsConfig":"{\"liveDurationInfinity\":true,\"liveBackBufferLength\":5,\"maxBufferLength\":10,\"manifestLoadingTimeOut\":1000,\"manifestLoadingMaxRetry\":10,\"manifestLoadingRetryDelay\":500}","autoplay":"true","unload":"true","threshold":"0.1","controls":"false","muted":"false","players":["socket.io","hls.js","hls","mp4"],"x":920,"y":4500,"wires":[[]]},{"id":"7a8c3f95b486b8f9","type":"ffmpeg","z":"44908087.da8e9","name":"","outputs":3,"cmdPath":"ffmpeg","cmdArgs":"[\"-loglevel\",\"quiet\",\"-rtsp_transport\",\"tcp\",\"-i\",\"rtsp://nvr:12345@192.168.1.95:554/h264Preview_01_sub\",\"-an\",\"-c:v\",\"copy\",\"-f\",\"mp4\",\"-movflags\",\"+frag_keyframe+empty_moov+default_base_moof\",\"pipe:1\"]","cmdOutputs":2,"killSignal":"SIGTERM","credentials":{},"x":630,"y":4640,"wires":[["4cc7afbbcc802243"],["4cc7afbbcc802243"],["30bbbddd32d43bde"]]},{"id":"71f89f0fcd720dc8","type":"inject","z":"44908087.da8e9","name":"restart","props":[{"p":"action","v":"{\"command\":\"restart\"}","vt":"json"}],"repeat":"","crontab":"00 06 * * *","once":false,"onceDelay":0.1,"topic":"","x":380,"y":4600,"wires":[["7a8c3f95b486b8f9"]]},{"id":"7ea407c1b15941c7","type":"ui_group","name":"Camera Deur","tab":"d9c0e801b43deec2","order":1,"disp":false,"width":"21","collapse":false,"className":""},{"id":"d9c0e801b43deec2","type":"ui_tab","name":"Camera Deur","icon":"fa-video-camera","order":27,"disabled":false,"hidden":false}]

[kevinGodell]

It could possibly be your input camera or your browser/device used to play the video.

So, the first thing to figure out is how long of durations the media segments are coming from the camera. Based on your settings, you are copying the video with ffmpeg, which puts you at the mercy of whatever the camera feed gives you. For example, I have 1 camera that cannot output media segments less than 4 seconds duration. And I have another cam that changes the durations of media segments based on if it is in day or night mode. Neither of the cams allow me to change those settings. This causes my playback to be a little erratic at times. Maybe yours does something similar. There can be a fix for that, but first we need to know what it is doing.

I have updated the flow you shared and then added a function node that can show some details that we need to get.

[{"id":"372df3f2323e3a3f","type":"inject","z":"9f566d31b9e867ee","name":"start","props":[{"p":"action","v":"{\"command\":\"start\"}","vt":"json"}],"repeat":"","crontab":"","once":false,"onceDelay":0.1,"topic":"","x":270,"y":260,"wires":[["933322a61f8b7c41"]]},{"id":"933322a61f8b7c41","type":"ffmpeg","z":"9f566d31b9e867ee","name":"","outputs":3,"cmdPath":"ffmpeg","cmdArgs":"[\"-loglevel\",\"quiet\",\"-rtsp_transport\",\"tcp\",\"-i\",\"rtsp://nvr:12345@192.168.1.95:554/h264Preview_01_sub\",\"-an\",\"-c:v\",\"copy\",\"-f\",\"mp4\",\"-movflags\",\"+frag_keyframe+empty_moov+default_base_moof\",\"pipe:1\"]","cmdOutputs":2,"killSignal":"SIGTERM","x":510,"y":340,"wires":[["4a2a1286f0a6d95a"],["4a2a1286f0a6d95a"],["a8a6b62e943dad43"]]},{"id":"d49b9764533a74e3","type":"inject","z":"9f566d31b9e867ee","name":"stop","props":[{"p":"action","v":"{\"command\":\"stop\"}","vt":"json"}],"repeat":"","crontab":"","once":false,"onceDelay":0.1,"topic":"","x":270,"y":380,"wires":[["933322a61f8b7c41"]]},{"id":"71f89f0fcd720dc8","type":"inject","z":"9f566d31b9e867ee","name":"restart","props":[{"p":"action","v":"{\"command\":\"restart\"}","vt":"json"}],"repeat":"","crontab":"00 06 * * *","once":false,"onceDelay":0.1,"topic":"","x":260,"y":300,"wires":[["933322a61f8b7c41"]]},{"id":"4a2a1286f0a6d95a","type":"mp4frag","z":"9f566d31b9e867ee","name":"","outputs":2,"basePath":"ui_mp4frag_01","serveHttp":"true","serveIo":"true","hlsPlaylistSize":"6","hlsPlaylistExtra":"2","autoStart":"false","preBuffer":1,"timeLimit":10000,"repeated":"false","statusData":"all","x":770,"y":320,"wires":[["dc113a743b091d48"],[]]},{"id":"a8a6b62e943dad43","type":"debug","z":"9f566d31b9e867ee","name":"stderr","active":false,"tosidebar":true,"console":false,"tostatus":false,"complete":"true","targetType":"full","statusVal":"","statusType":"auto","x":700,"y":480,"wires":[]},{"id":"dc113a743b091d48","type":"function","z":"9f566d31b9e867ee","name":"status","func":"const { status, payload } = msg;\n\n// most frequent, handle segment status first\nif (status === 'segment') {\n\n    let { sequence, duration, totalDuration, byteLength, totalByteLength, keyframe } = payload;\n\n    // duration of current segment\n    duration = duration.toFixed(1);\n\n    // total duration of all segments\n    totalDuration = totalDuration.toFixed(1);\n\n    // byte length of current segment\n    byteLength = (byteLength / 1024).toFixed(0);\n\n    // total byte length of all segments\n    totalByteLength = (totalByteLength / 1024).toFixed(0);\n\n    // indicate if current segment contains keyframe\n    keyframe = keyframe ? 1 : 0;\n\n    const text = `${duration}/${totalDuration}(s), ${byteLength}/${totalByteLength}(KB), ${keyframe}, ${sequence}`;\n\n    node.status({ fill: 'green', shape: 'dot', text });\n\n    return null;\n\n}\n\n// 2nd most frequent, pass this msg along to ui_mp4frag\nif (status === 'playlist' || status === 'reset') {\n\n    return msg;\n\n}\n\n// least frequent, initialization fragment is parsed\nif (status === 'initialized') {\n\n    const { mime/*, videoCodec, audioCodec*/ } = payload;\n\n    // const text = `vc: ${videoCodec}, ac: ${audioCodec}`;\n\n    node.status({ fill: 'green', shape: 'dot', text: mime });\n\n    return null;\n\n}","outputs":1,"noerr":0,"initialize":"","finalize":"","libs":[],"x":1010,"y":300,"wires":[["91780c25f9e1edae"]]},{"id":"91780c25f9e1edae","type":"ui_mp4frag","z":"9f566d31b9e867ee","name":"Camera Deur","group":"7ea407c1b15941c7","order":1,"width":"21","height":"11","readyPoster":"data:image/png;base64,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","errorPoster":"data:image/png;base64,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","hlsJsConfig":"{\"liveDurationInfinity\":true,\"liveBackBufferLength\":5,\"maxBufferLength\":10,\"manifestLoadingTimeOut\":1000,\"manifestLoadingMaxRetry\":10,\"manifestLoadingRetryDelay\":500}","autoplay":"true","unload":"true","threshold":"0.1","controls":"false","muted":"false","players":["socket.io","hls.js","hls","mp4"],"x":1210,"y":300,"wires":[[]]},{"id":"7ea407c1b15941c7","type":"ui_group","name":"Camera Deur","tab":"d9c0e801b43deec2","order":1,"disp":false,"width":"21","collapse":false,"className":""},{"id":"d9c0e801b43deec2","type":"ui_tab","name":"Camera Deur","icon":"fa-video-camera","order":27,"disabled":false,"hidden":false}]

Let me know if you can get that segment duration info or make a screenshot of the function node while the camera feed is running. Also, please let me know if you are using mobile phone or desktop and which browsers.

[edje11]

Hi Kevin,
Thanks again for your prompt and quick reply.
My other 3 camera's (Annke) are working fine and don't have a delay that will increase overtime.
Most likely the problem is indeed coming from this Reolink doorbel.
Here is the requested screenshot:

The problem of the increasing delay occurs on all my devices, however testing is always be done on my win10 desktop in Mozilla firefox 111.0.1

[kevinGodell]

Let's will break down what each of the numbers mean for 4.0/32.0(s), 95/834(KB), 1, 21

4 - the most recent segment was of 4 seconds durations.
32 - there is a total of 32 seconds of video buffered (in memory).
95 - the most recent was 95 kilobytes
834 - total of 834 kilobytes of video buffered (in memory).
1 - current segment contains a keyframe/iframe
21 - 21st segment to be received through mp4frag from ffmpeg

Those aforementioned values totals are based on adding up the numbers in mp4frag settings under buffered segments size/extra. From your shared flow, I will conclude that your are still using 6 + 2.

(6+2) x 4 = 32. It looks like all of you segments were approximately 4 seconds duration, which is good because it means they are pretty consistent. I wonder if your segment duration changes with the lighting of color vs black&white modes. Further testing would be needed to verify if it fluctuates.

If the duration is 4 seconds for a segment, then the real-time viewing will be +4 seconds behind realtime due to the nature of when ffmpeg spits out the newest segment and mp4frag can parse it.

In order to keep the latency low, there is a newer feature of ffmpeg that allows you to still benefit from the low cpu load option of doing stream copying, while packaging the segments into shorter durations.

If you change the movflags from +frag_keyframe+empty_moov+default_base_moof to +frag_every_frame+empty_moov+default_base_moof, it will cause ffmpeg to fragment every single frame into its own segment. That would be overkill and not function well, so we must add an extra parameter to limit the size of the segment so that it is not so small.

"-movflags",
"+frag_every_frame+empty_moov+default_base_moof",
"-min_frag_duration",
"1000000",

This will give you segments that are close to a duration of 1 second. If you do this, you will notice that the value on that function node will change from 1 to 0 to indicate if that segment had a keyframe or not. If you are doing recordings with this data, then more things will have to be considered. Let me know if this helps.

[edje11]

Hi Kevin,
Thanks again for you extensive explanation.
I have changed the movflags settings and the numbers are looking now different as expected.

The initial delay dropped from 5 to 4 seconds showing the video in the dashboard.
Cpu load (pi-3B+) increased from 1.6 to 2% for ffmpeg so that's nothing.
To test if the delay will increase overtime needs sometime to test and also the night mode of the camera takes some time to test. Night mode is also in color, but I will test.

Will keep you posted.

[krambriw]

Just a question related; If you have, in this case 8,8 seconds of video buffered in memory, is that history or new video waiting to be presented/viewed? If new video, doesn't that mean you are 8.8 seconds late in relation to live view?

[kevinGodell]

@krambriw

in this case 8,8 seconds of video buffered in memory, is that history or new video waiting to be presented/viewed

It is history AND video waiting to be viewed. I guess that as soon as we receive a single segment, it is already in the history, even if it is just 1 second in the past.

@krambriw

If new video, doesn't that mean you are 8.8 seconds late in relation to live view

That also depends on the front end / client side video player. hls.js or native HLS usually need to consume more than a single segment before it can start playing video, as opposed to the hackey socket .io streamer that can play it right away and does not need a list of segments. We are always late, but try to be less late when the conditions are just right, such as small video size, fast internet, good browser with plenty of ram, etc.

If using the buffer for initiating a recording of recently past video, then the size/extra can be altered to keep a reasonable amount of video available for that. Since i am doing 24/7 recording vs motion detection recording, I have kept my size at 4 and extra at 1 to not keep as much old video. This still allows the native hls on my old iphone 6s safari browser to play the video smoothly.

I guess this stuff could be called recent-past vs live, but then it wouldn't sound as nice.