You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
Joachim Bauch 4530b91434
Merge pull request #153 from strukturag/ci-go-1.17
7 days ago
.github CI: Also test with Golang 1.17 7 days ago
client No need to include trailing newline in log messages. 4 months ago
dist/init/systemd add systemd to docs 1 year ago
docker Update Docker components (#78) 5 months ago
proxy Close publishers/subscribers if proxy session is closed. 1 month ago
scripts Fix various issues found by golangci-lint. 5 months ago
server Add prometheus metrics. 1 month ago
.dockerignore Don't include bin and vendor folders in Docker build context. 5 months ago
.gitignore Add proxy service implementation. 1 year ago
CHANGELOG.md Update changelog for 0.3.0 3 months ago
Dockerfile python3 is no longer needed for building since #29 has been merged. 5 months ago
LICENSE Initial commit of the OpenSource version. 1 year ago
Makefile adjust easyjson for multiarch builds 3 months ago
README.md Add coverage status badge. 5 months ago
api_backend.go Fix various issues found by golangci-lint. 5 months ago
api_backend_test.go Switch dependency tracking to Go modules. 5 months ago
api_proxy.go Fix various issues found by golangci-lint. 5 months ago
api_signaling.go Add feature flag "simulcast". 3 months ago
api_signaling_test.go Switch dependency tracking to Go modules. 5 months ago
backend_client.go Simplify request creation when getting capabilities. 3 months ago
backend_client_test.go Only include body in 307/308 redirects if going to same host. 3 months ago
backend_configuration.go Add more metrics on rooms / sessions / backends. 1 month ago
backend_configuration_stats_prometheus.go Add more metrics on rooms / sessions / backends. 1 month ago
backend_configuration_test.go Add more metrics on rooms / sessions / backends. 1 month ago
backend_server.go Add prometheus metrics. 1 month ago
backend_server_test.go Force HTTPS for backend connections in old-style configurations. 3 months ago
client.go Add more metrics on rooms / sessions / backends. 1 month ago
client_stats_prometheus.go Add more metrics on rooms / sessions / backends. 1 month ago
clientsession.go Stop publishers if session is no longer allowed to publish. 2 months ago
clientsession_test.go Use subtests for testing with different values. 4 months ago
concurrentmap.go Switch dependency tracking to Go modules. 5 months ago
concurrentmap_test.go Switch dependency tracking to Go modules. 5 months ago
continentmap.go Fix various issues found by golangci-lint. 5 months ago
deferred_executor.go Switch dependency tracking to Go modules. 5 months ago
deferred_executor_test.go Switch dependency tracking to Go modules. 5 months ago
docker-compose.yml Update Docker components (#78) 5 months ago
geoip.go Add method "IsValidContinent". 2 months ago
geoip_test.go Add method "IsValidContinent". 2 months ago
gnatsd.conf Initial commit of the OpenSource version. 1 year ago
go.mod Bump github.com/nats-io/nats.go from 1.11.0 to 1.12.1 2 weeks ago
go.sum Bump github.com/nats-io/nats.go from 1.11.0 to 1.12.1 2 weeks ago
hub.go Fixup: don't decrement virtual sessions twice. 1 month ago
hub_stats_prometheus.go Add more metrics on rooms / sessions / backends. 1 month ago
hub_test.go Force HTTPS for backend connections in old-style configurations. 3 months ago
janus_client.go Fix various issues found by golangci-lint. 5 months ago
lru.go Switch dependency tracking to Go modules. 5 months ago
lru_test.go Switch dependency tracking to Go modules. 5 months ago
mcu_common.go Fix various issues found by golangci-lint. 5 months ago
mcu_common_test.go Lint prometheus stats. 1 month ago
mcu_janus.go Also track subscriber streams by type. 1 month ago
mcu_janus_test.go Lint prometheus stats. 1 month ago
mcu_proxy.go Add prometheus metrics. 1 month ago
mcu_proxy_test.go Lint prometheus stats. 1 month ago
mcu_stats_prometheus.go Also track subscriber streams by type. 1 month ago
mcu_test.go Support defining maximum bandwidths at diferent levels. 5 months ago
natsclient.go Also track subscriber streams by type. 1 month ago
natsclient_loopback.go Ignore room NATS messages that were sent before the room was joined. 3 months ago
natsclient_loopback_test.go Simplify loopback NATS client. 4 months ago
natsclient_test.go Don't start test until receiving goroutine is ready. 3 months ago
notifier.go Support multiple waiters for the same key. 4 months ago
notifier_test.go Support multiple waiters for the same key. 4 months ago
pool.go Switch dependency tracking to Go modules. 5 months ago
pool_test.go Switch dependency tracking to Go modules. 5 months ago
proxy.conf.in Assume one Janus server is only used by one signaling server / -proxy. 4 months ago
room.go Add more metrics on rooms / sessions / backends. 1 month ago
room_stats_prometheus.go Add more metrics on rooms / sessions / backends. 1 month ago
room_test.go Always use auth userid in ping requests. 4 months ago
roomsessions.go Switch dependency tracking to Go modules. 5 months ago
roomsessions_builtin.go Switch dependency tracking to Go modules. 5 months ago
roomsessions_builtin_test.go Switch dependency tracking to Go modules. 5 months ago
roomsessions_test.go Fix various issues found by golangci-lint. 5 months ago
server.conf.in Support continent mapping overrides. 2 months ago
session.go Switch dependency tracking to Go modules. 5 months ago
session_test.go Switch dependency tracking to Go modules. 5 months ago
stats_prometheus.go Add prometheus metrics. 1 month ago
stats_prometheus_test.go Add more metrics on rooms / sessions / backends. 1 month ago
testclient_test.go Add method to get pending messages from testclient. 3 months ago
testutils_test.go Add tests for regular NATS client. 4 months ago
virtualsession.go Send more responses through the session to handle disconnected / resuming clients. 4 months ago
virtualsession_test.go Notify new clients about current flags of virtual sessions. 3 months ago

README.md

Spreed standalone signaling server

Build Status Coverage Status

This repository contains the standalone signaling server which can be used for Nextcloud Talk (https://apps.nextcloud.com/apps/spreed).

See https://nextcloud-talk.readthedocs.io/en/latest/standalone-signaling-api-v1/ for further information on the API of the signaling server.

Building

The following tools are required for building the signaling server.

  • git
  • go >= 1.13
  • make

All other dependencies are fetched automatically while building.

$ make build

or on FreeBSD

$ gmake build

Afterwards the binary is created as bin/signaling.

Configuration

A default configuration file is included as server.conf.in. Copy this to server.conf and adjust as necessary for the local setup. See the file for comments about the different parameters that can be changed.

Running

The signaling server connects to a NATS server (https://nats.io/) to distribute messages between different instances. See the NATS documentation on how to set up a server and run it.

Once the NATS server is running (and the URL to it is configured for the signaling server), you can start the signaling server.

$ ./bin/signaling

By default, the configuration is loaded from server.conf in the current directory, but a different path can be passed through the --config option.

$ ./bin/signaling --config /etc/signaling/server.conf

Running as daemon

systemd

Create a dedicated group and user:

sudo groupadd --system signaling
sudo useradd --system \
    --gid signaling \
    --shell /usr/sbin/nologin \
    --comment "Standalone signaling server for Nextcloud Talk." \
    signaling

Copy server.conf.in to /etc/signaling/server.conf and fix permissions:

sudo chmod 600 /etc/signaling/server.conf
sudo chown signaling: /etc/signaling/server.conf

Copy dist/init/systemd/signaling.service to /etc/systemd/system/signaling.service (adjust abs. path in ExecStart to match your binary location!)

Enable and start service:

systemctl enable signaling.service
systemctl start signaling.service

Running with Docker

Docker Compose

You will likely have to adjust the Janus command line options depending on the exact network configuration on your server. Refer to Setup of Janus and the Janus documentation for how to configure your Janus server.

Copy server.conf.in to server.conf and adjust it to your liking.

If you're using the docker-compose.yml configuration as is, the MCU Url must be set to ws://localhost:8188, the NATS Url must be set to nats://localhost:4222, and TURN Servers must be set to turn:localhost:3478?transport=udp,turn:localhost:3478?transport=tcp.

docker-compose build
docker-compose up -d

Setup of NATS server

There is a detailed description on how to install and run the NATS server available at http://nats.io/documentation/tutorials/gnatsd-install/

You can use the gnatsd.conf file as base for the configuration of the NATS server.

Setup of Janus

A Janus server (from https://github.com/meetecho/janus-gateway) can be used to act as a WebRTC gateway. See the documentation of Janus on how to configure and run the server. At least the VideoRoom plugin and the websocket transport of Janus must be enabled.

The signaling server uses the VideoRoom plugin of Janus to manage sessions. All gateway details are hidden from the clients, all messages are sent through the signaling server. Only WebRTC media is exchanged directly between the gateway and the clients.

Edit the server.conf and enter the URL to the websocket endpoint of Janus in the section [mcu] and key url. During startup, the signaling server will connect to Janus and log information of the gateway.

The maximum bandwidth per publishing stream can also be configured in the section [mcu], see properties maxstreambitrate and maxscreenbitrate.

Use multiple Janus servers

To scale the setup and add high availability, a signaling server can connect to one or multiple proxy servers that each provide access to a single Janus server.

For that, set the type key in section [mcu] to proxy and set url to a space-separated list of URLs where a proxy server is running.

Each signaling server that connects to a proxy needs a unique token id and a public / private RSA keypair. The token id must be configured as token_id in section [mcu], the path to the private key file as token_key.

Setup of proxy server

The proxy server is built with the standard make command make build as bin/proxy binary. Copy the proxy.conf.in as proxy.conf and edit section [tokens] to the list of allowed token ids and filenames of the public keys for each token id. See the comments in proxy.conf.in for other configuration options.

When the proxy process receives a SIGHUP signal, the list of allowed token ids / public keys is reloaded. A SIGUSR1 signal can be used to shutdown a proxy process gracefully after all clients have been disconnected. No new publishers will be accepted in this case.

Setup of frontend webserver

Usually the standalone signaling server is running behind a webserver that does the SSL protocol or acts as a load balancer for multiple signaling servers.

The configuration examples below assume a pre-configured webserver (nginx or Apache) with a working HTTPS setup, that is listening on the external interface of the server hosting the standalone signaling server.

After everything has been set up, the configuration can be tested using curl:

$ curl -i https://myserver.domain.invalid/standalone-signaling/api/v1/welcome
HTTP/1.1 200 OK
Date: Thu, 05 Jul 2018 09:28:08 GMT
Server: nextcloud-spreed-signaling/1.0.0
Content-Type: application/json; charset=utf-8
Content-Length: 59

{"nextcloud-spreed-signaling":"Welcome","version":"1.0.0"}

nginx

Nginx can be used as frontend for the standalone signaling server without any additional requirements.

The backend should be configured separately so it can be changed in a single location and also to allow using multiple backends from a single frontend server.

Assuming the standalone signaling server is running on the local interface on port 8080 below, add the following block to the nginx server definition in /etc/nginx/sites-enabled (just before the server definition):

upstream signaling {
    server 127.0.0.1:8080;
}

To proxy all requests for the standalone signaling to the correct backend, the following location block must be added inside the server definition of the same file:

location /standalone-signaling/ {
    proxy_pass http://signaling/;
    proxy_http_version 1.1;
    proxy_set_header Host $host;
    proxy_set_header X-Real-IP $remote_addr;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
}

location /standalone-signaling/spreed {
    proxy_pass http://signaling/spreed;
    proxy_http_version 1.1;
    proxy_set_header Upgrade $http_upgrade;
    proxy_set_header Connection "Upgrade";
    proxy_set_header Host $host;
    proxy_set_header X-Real-IP $remote_addr;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
}

Example (e.g. /etc/nginx/sites-enabled/default):

upstream signaling {
    server 127.0.0.1:8080;
}

server {
    listen 443 ssl http2;
    server_name myserver.domain.invalid;

    # ... other existing configuration ...

    location /standalone-signaling/ {
        proxy_pass http://signaling/;
        proxy_http_version 1.1;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
    }

    location /standalone-signaling/spreed {
        proxy_pass http://signaling/spreed;
        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection "Upgrade";
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
    }
}

Apache

To configure the Apache webservice as frontend for the standalone signaling server, the modules mod_proxy_http and mod_proxy_wstunnel must be enabled so WebSocket and API backend requests can be proxied:

$ sudo a2enmod proxy
$ sudo a2enmod proxy_http
$ sudo a2enmod proxy_wstunnel

Now the Apache VirtualHost configuration can be extended to forward requests to the standalone signaling server (assuming the server is running on the local interface on port 8080 below):

<VirtualHost *:443>

    # ... existing configuration ...

    # Enable proxying Websocket requests to the standalone signaling server.
    ProxyPass "/standalone-signaling/"  "ws://127.0.0.1:8080/"

    RewriteEngine On
    # Websocket connections from the clients.
    RewriteRule ^/standalone-signaling/spreed/$ - [L]
    # Backend connections from Nextcloud.
    RewriteRule ^/standalone-signaling/api/(.*) http://127.0.0.1:8080/api/$1 [L,P]

    # ... existing configuration ...

</VirtualHost>

Caddy

v1

Caddy (v1) configuration:

myserver.domain.invalid {
  proxy /standalone-signaling/ http://127.0.0.1:8080 {
    without /standalone-signaling
    transparent
    websocket
  }
}

v2

Caddy (v2) configuration:

myserver.domain.invalid {
  route /standalone-signaling/* {
    uri strip_prefix /standalone-signaling
    reverse_proxy http://127.0.0.1:8080
  }
}

Setup of Nextcloud Talk

Login to your Nextcloud as admin and open the additional settings page. Scroll down to the "Talk" section and enter the base URL of your standalone signaling server in the field "External signaling server". Please note that you have to use https if your Nextcloud is also running on https. Usually you should enter https://myhostname/standalone-signaling as URL.

The value "Shared secret for external signaling server" must be the same as the property secret in section backend of your server.conf.

If you are using a self-signed certificate for development, you need to uncheck the box Validate SSL certificate so backend requests from Nextcloud to the signaling server can be performed.

Benchmarking the server

A simple client exists to benchmark the server. Please note that the features that are benchmarked might not cover the whole functionality, check the implementation in src/client for details on the client.

To authenticate new client connections to the signaling server, the client starts a dummy authentication handler on a local interface and passes the URL in the hello request. Therefore the signaling server should be configured to allow all backend hosts (option allowall in section backend).

The client is not compiled by default, but can be using the client target:

$ make client

Usage:

$ ./bin/client
Usage of ./bin/client:
  -addr string
        http service address (default "localhost:28080")
  -config string
        config file to use (default "server.conf")
  -maxClients int
        number of client connections (default 100)

Running multiple signaling servers

IMPORTANT: This is considered experimental and might not work with all functionality of the signaling server, especially when using the Janus integration.

The signaling server uses the NATS server to send messages to peers that are not connected locally. Therefore multiple signaling servers running on different hosts can use the same NATS server to build a simple cluster, allowing more simultaneous connections and distribute the load.

To set this up, make sure all signaling servers are using the same settings for their session keys and the secret in the backend section. Also the URL to the NATS server (option url in section nats) must point to the same NATS server.

If all this is setup correctly, clients can connect to either of the signaling servers and exchange messages between them.