Before you start
You need an active video channel with SRT enabled and an encoder that supports SRT. To test reception, also prepare a compatible receiver, such as VLC or FFmpeg built with SRT support. SRT ingest and output are included with video plans; plan limits and transfer allowances still apply.
- Get the current connection details from the panel and ask support about any missing field.
- Prepare a test signal with video, audio and a visible clock to assess the full path.
- Check that the network permits UDP traffic to the assigned host and port. A working RTMP connection does not confirm that SRT is allowed.
1. Separate publishing, playback and SRT connection mode
| Connection | Your device | Details to use |
|---|---|---|
| Publish to XtreamCast | Encoder in caller mode; the server is the listener. | SRT ingest URL and the channel publishing credential. |
| Receive from XtreamCast | Receiver in caller mode; the server is the listener. | The channel SRT playback URL. |
Caller identifies the endpoint that starts the connection; it does not necessarily mean the sender. Copy input and output addresses separately. Playback may use a dedicated port or a shared port with a stream ID, depending on the channel assignment. Do not invent a port or add a stream ID to an address supplied without one.
2. Copy the complete URL and protect the stream ID
Current ingest connections using a stream ID identify the channel and carry a publishing token. This example shows the structure with fictional placeholders; it is not a working address:
srt://HOST:PORT?mode=caller&streamid=VHOST%2FAPP%2FSTREAM%3Ftoken%3DTOKENIf the encoder has a separate Stream ID field, its logical format is shown below. Use the value in the format your encoder expects; %2F, %3F and %3D in a URL represent encoded characters. Do not encode the value twice.
VHOST/APP/STREAM?token=TOKENThe publishing URL and token are confidential. Keep them out of public websites, repositories, screenshots and unredacted tickets. If exposed, request replacement credentials and update your encoders. Token authentication does not prove that the transport is encrypted.
Authentication formats can vary across servers. Always use current panel details. Do not add passphrase, pbkeylen or parameters copied from another provider. If you require transport encryption, confirm availability and the configuration of both endpoints with support before streaming.
3. Configure the encoder and send a test signal
- Select an SRT output in the encoder. Field names depend on the application and version.
- Set caller mode and paste the complete panel URL, or enter host, port and stream ID separately if the encoder requires it.
- Use codecs, resolution and bitrate compatible with your channel. Confirm the profile with support if a destination or player needs specific settings.
- Adjust the SRT latency buffer for the network you will use. Check the encoder units: latency fields are not expressed the same way in every application.
- Start the output and check that the panel receives video and audio. If there is no signal, stop the test and use the troubleshooting table.
SRT can recover packets within its time window. A buffer that is too short reduces the recovery margin; a larger buffer adds delay. SRT does not replace sufficient connection capacity or recover a prolonged disconnection.
4. Check ingest and playback separately
- Open the channel player or HLS URL to confirm that ingest reached the server. This playback path has its own buffer.
- Open the SRT playback URL in a compatible receiver. Do not use the secret publishing URL for this check.
- Check video, audio, synchronization and continuity. Repeat using the networks and devices planned for the event.
- To estimate latency, compare a clock visible in the source with the received picture under the same conditions. Record the encoder, buffer, network, receiver and test time.
- Rehearse link loss and recovery on a test channel. Decide who switches to a backup; do not assume automatic failover.
These are validation steps for you to run on your channel. They are not published measurements of latency, availability or packet recovery.
Troubleshooting common problems
| Symptom | What to check |
|---|---|
| Connection fails or times out | Host, port, caller mode, active channel and permitted UDP traffic. Confirm that SRT is enabled on the channel and server. |
| Connects, then immediately closes | Token, complete stream ID and the correct ingest or playback URL. Copy fresh values from the panel without extra spaces. |
| Connected, but no video or audio | Active encoder output, compatible codecs and track selection. Test ingest first, then each output. |
| Stuttering or interrupted picture | Bitrate versus available upload, network loss and SRT buffer. Reduce bitrate and measure again before increasing resolution. |
| More delay than expected | Capture, encoder, SRT buffer, distance and receiver. Do not confuse transport delay with total HLS player latency. |
When contacting support, provide the channel ID, date and time with time zone, encoder and version, network used and observed error. Redact tokens and passwords from screenshots and logs.
Limits to plan for
Zero loss, recovery of every packet and fixed latency are not guaranteed. Encryption requires configuration confirmation. FEC, bonding multiple connections and automatic backup switching do not follow from SRT support: if your production needs them, validate the equipment and service scope separately.