Trying to solve a perplexing issue w/streaming audio over websockets. We are using Nexmo (Twilio competitor) which enables bidirectional streaming of call audio over websockets. Nexmo connects to our websocket server and starts sending 16khz sampled audio frames of length 640 bytes each.
Everything was working great until recently the websocket audio suddenly started dropping clumps of frames, resulting in gaps in the audio.
But the most interesting thing is the following:
When Nexmo connects directly to our digitalocean vps, frames are dropped
When Nexmo connects via an ngrok tunnel, everything starts working again
Any ideas on where to look for a real solution would be awesome.
Make certain the process which is receiving the websocket traffic has a separate thread just to handle this traffic ... any system will drop traffic if its too busy with other tasks ... if your receiving end has some event loop its maintaining while getting preempted by incoming websocket interrupts you will drop packets
I did a project where receiving end was the browser which was running an event loop to perform the audio rendering while simultaneously also handling the websocket traffic - not a good idea since the critical portions of this event loop must not be allowed to get preempted ... I had to create a webworker process on browser side to handle all the websocket traffic to then populate a circular audio buffer ... this webworker was viewed as a client by the browser event loop which was rendering the audio yet was now permitted to never get preempted by incoming traffic ... only when browser event loop reached its lull period did it then request to retrieve another gulp of data buffered up by webworker audio buffer queue
Related
I have an interesting problem, in short: how to share information between threads in NodeJS (12+).
The tech stack - in short also:
A remote/online streaming server, what producing an MP4 live stream
A client application what only consumes live view through RTSP over HTTP
A small NodeJS based application to get the MP4, transform it and pipe it back to the client.
.
The modules what I use:
NodeJS 12+
Request/fetch/https module
Express module
Stream module
The story:
I have an application, what has a gateway/relay role between two different system. One provide a live media stream (simple MP4(h264) stream) and another one supposed to consume it as RTSP over HTTP. The weird part is, the consumer client does not behave like any other player (like VLC or a webplayer), sometime - seemingly randomly - resend the request, sometime close the current request and resend it. So direct pipe not really working for this use-case.
I made a worker (from worker_threads), what hold a readable stream object, and when the client hit the request, I start populate the MP4 stream into the readable object in the worker, so even if the stream is does get a close or resend, it will not break the live media stream consuming process.
And wherever the client connect, I just would like to pipe the readable object for it.
Originally, I though a simple pipe from like request/fetch/http.get or FFMPEG would be enough, but the client could call the call between 3 seconds and 2 minutes.
.
So, my questions are, what could be the best solution, to pass back the data from the worker to the main and let reach the HTTP routing?
I had some idea like:
I know, I can have my own channel between the threads and can pass back-and-forth information, but waiting for message and keep up the process does block the app, as far as I know (worker.on('message', (stuff) => {});).
Using Socket.io to pass data back from the worker, populate the readable in the main, and pipe the readable at http level (fake shared object basically)
Creating a secondary http server what offer the media stream, then i will just relay this into the response (e.g.: gatewaying/proxying)
Looking up some proxy solution where I can just simply redirect and reshape thing, like the input mp4 transforms into RTSP stream and pipe it to the consumer response
Should I just "remember" to the active stream, and if its streamed by the remote server, always just using the same url, passing to FFMPEG and continue piping to the res?
Note:
I setted up all the headers to keep alive the connection, but seems the client software act as-is.
By default its using RTSP and RTP/TCP to consume video stream, but has option for RTSP over http.
Probably I overlook some trivial task for serving RTSP video from a remote live MP4, but I did not found any good example or source anywhere (everywhere the same 3 article re-shared basically)
I did not found any similar question, nor article anywhere (but checked out the nodejs ffmpeg play video at specific time and stream it to client).
I am working on frontend application, where front end application sends video (live) frames to backend server, backend server processes the video frames and sends the data(text) back to front end application via websocket connection. On backend websocket server side, whenever client sends video frames, these frames are added to the queue along with connection id. Is it possible to write a method which polls the queue, process the frames and generate the text data for specific client which will be sent to the client. This method should have one instance only and should be shared by all websocket connections. The method uses the shared variables and should have exclusive access. I can't mix frames from different client to generate text data.
If I write a producer then each client connection will have separate producer and these producers will use shared variable which is not desired. Right now I wrote websocket connection without consumer-producer and put lock around the critical code. But it doesn't seem good solution. Backend is in python and front end in reactjs. Any suggestions?
I am using the ws Node.js package to create a simple WebSocket client connection to a server that is sending hundreds of messages per second. Even with a simple onMessage handler that just console.logs incoming messages, the client cannot keep up. My understanding is that this is referred to as backpressure, and incoming messages may start piling up in a network buffer on the client side, or the server may throttle the connection or disconnect all-together.
How can I monitor backpressure, or the network buffer from the client side? I've found several articles speaking about this issue from the perspective of the server, but I have no control over the server and need to know just how slow is my client?
So you don't have control over the server and want to know how slow your client is.(seems like you already have read about backpressure). Then I can only think of using a stress tool like artillery
Check this blog, it might help you setting up a benchmarking scenario.
https://ma.ttias.be/benchmarking-websocket-server-performance-with-artillery/
Add timing metrics to your onMessage function to track how long it takes to process each message. You can also use RUM instrumentation like from the APM providers -- NewRelic or Appdynamics for paid options or you could use free tier of Google Analytics timing.
If you can, include a unique identifier for correlation between the client and server for each message sent.
Then you can correlate for a given window how long a message took to send from the server and how long it spent being processed by the client.
You can't get directly to the network socket buffer associated with your websocket traffic since you're inside the browser sandbox. I checked the WebSocket APIs and there's no properties that expose receive buffer information.
If you don't have control over the server, you are limited. But you could try some client tricks to simulate throttling.
This heavily assumes you don't mind skipping messages.
One approach would be to enable the socket, start receiving events and set your own max count in a in-memory queue/array. Once you reach a full queue, turn off the socket. Process enough of the queue, then enable the socket again.
This has high cost to disable/enable the socket, as well as the loss of events, but at least your client will not crash.
Once your client is not crashing, you can put some additional counts on timestamp and the queue size to determine the threshold before the client starts crashing.
I'm writing an audio streaming server - similar to Icecast, and I'm running into a problem with streaming audio files. Proxying audio works fine (an audio source connects and sends audio in real time, which is then transmitted to clients over HTTP), but when I try to stream an audio file it goes by to quickly - clients end up with the entire audio file within their local buffer. I want them to only have a few 10s of seconds in their local buffer.
Essentially, how can I slow down the sending of an audio file over HTTP?
The files are all MP3. I've managed to get it pretty much working by experimenting with hardcoded thread delays etc... but that's not a sustainable solution.
If you're sticking with http you could use chunked transfer encoding and delay sending the packets/chunks. This would indeed be something similar to hardcoded thread::sleep but you could use an event loop to determine when to send the next chunk instead of pausing the thread.
You might run into timing issues though, maybe your sleep logic is causing longer delays than the runtime of the song. YouTube has similar logic to what you're talking about. It looks like they break videos into multiple http requests and the frontend client requests a new chunk when the buffer is too small. Breaking the file into multiple http body requests and then reassembling them at the client might have the characteristics you're looking for.
You could simply implement the http Range header and allow the client to only request a specific Range of the mp3 file. https://developer.mozilla.org/en-US/docs/Web/HTTP/Range_requests
The easiest method (by far) would be to have the client request chunks of the audio file on demand. std::net::TcpStream (which is what you said you're using) doesn't have a method to throttle the transfer rate, so you don't have many options to limit streaming backend short of using hard-coded thread delays.
As an example, you can have your client store a segment of audio, and when the user listening to the audio reaches a certain point before the end of the segment (or skips ahead), the client makes a request to the server to fetch the relevant segment.
This is similar to how real-world streaming services (like Youtube) work, because as you said, it would be a bad idea to store the entire file client-side.
I'm new to socket programming and I need to implement a UDP based rateless file transmission system to verify a scheme in my research. Here is what I need to do:
I want a server S to send a file to a group of peers A, B, C.., etc. The file is divided into a number of packets. At the beginning, peers will send a Request message to the server to initialize transmission. Whenever S receives a request from a client, it ratelessly transmit encoded packets(how to encode is done by my design, the encoding itself has the erasure-correction capability, that's why I can transmit ratelessly via UDP) to that client. The client keeps collecting packets and try to decode them. When it finally decodes all packets and re-construct the file successfully, it sends back a Stop message to the server and S will stop transmitting to this client.
Peers request the file asynchronously (they may request the file at different time). And the server will have to be able to concurrently serve multiple peers. The encoded packets for different clients are different (they are all encoded from the same set source packets, though).
Here is what I'm thinking about the implementation. I have not much experience with unix network programming though, so I'm wondering if you can help me assess it, and see if it is possible or efficient.
I'm gonna implement the server as a concurrent UDP server with two socket ports(similar to TFTP according to the UNP book). One is to receive controlling messages, as in my context it is for the Request and Stop messages. The server will maintain a flag (=1 initially) for each request. When it receives a Stop message from the client, the flag will be set to 0.
When the serve receives a request, it will fork() a new process that use the second socket and port to send encoded packets to the client. The server keeps sending packets to the client as long as the flag is 1. When it turns to 0, the sending ends.
The client program is easy to do. Just send a Request, recvfrom() the server, progressively decode the file and send a Stop message in the end.
Is this design workable? The main concerns I have are: (1), is that efficient by forking multiple processes? Or should I use threads? (2), If I have to use multiple processes, how can the flag bit be known by the child process? Thanks for your comments.
Using UDB for file transfer is not best idea. There is no way for server or client to know if any packet has been lost so you would only know that during reconstruction assuming you have some mechanism (like counter) to detect lost packes. It would then be hard to request just one of those packets that got lost. And in the end you would have a code that would do what TCP sockets do. So I suggest to start with TCP.
Typical design of a server involves a listener thread that spawns a worker thread whenever there is a new client request. That new thread would handle communication with that particular client and then end. You should keep a limit of clients (threads) that are served simultaneously. Do not spawn a new process for each client - that is inefficient and not needed as this will get you nothing that you can't achieve with threads.
Thread programming requires carefulness so do not cut corners. Otherwise you will have hard time finding and diagnosing problems.
File transfer with UDP wil be fun :(
Your struct/class for each message should contain a sequence number and a checksum. This should enable each client to detect, and ask for the retransmission of, any missing blocks at the end of the transfer.
Where UDP might be a huge winner is on a local LAN. You could UDP-broadcast the entire file to all clients at once and then, at the end, ask each client in turn which blocks it has missing and send just those. I wish Kaspersky etc. would use such a scheme for updating all my local boxes.
I have used such a broadcast scheme on a CANBUS network where there are dozens of microControllers that need new images downloaded. Software upgrades take minutes instead of hours.