I found Wowza really nice piece of software, but of course as a beginner user I have some problems and questions about it.
- How Wowza treats ICY protocol data in shoutcast mode? It only separates audio data and metadata from ICY and put this stuff into RTMP protocol fields? Is audio data untouched (any encoding/decoding mechanism)? I have icecast server streaming audio in ACC+v2 encoding and I want to know if “support for HE-ACC” means that ACC+ in version 2 (not onlly v1) is supported.
I configured Wowza to re-stream this aac+v2 audio and it looks that everything works good, but I want to be sure that there is no quality loss in e.g. treating AAC+v2 as AAC+v1 stream (which probably should work but sound quality will be lower).
- Also when I configured this all ICY restreaming stuff i want to measure bandtwith differences between original ICY and RTMP and RTMPT. So I have about 30kbps ICY stream, it looks that RTMP is almost the same (30 kbps) but I have really strange results for RTMPT (about 120kbps for the same stream!).
I call RTMP by url:
and RTMPT by url:
I dumped some packets using tcpdump/wireshark and I saw that every request in RTMPT communication is made in new connection! So it looks that keep-alive is ignored in flash client or something?. I am using this example shoutcast flash player distributed with wowza. Keep-Alive is turned on server side in Wowza (I also see “Connection: Keep-Alive” in dumped HTTP packeds in response from Wowza). Am I doing something wrong? This 120kbps bandtwidth is not acceptable.
- I saw that I can define my own modules in Wowza, really cool! I want to create module which will work with shoutcast re-streaming and will add some “intro sound” for every user which connects to Wowza. It is possible? I saw method “addAudioData(byte data, int offset, int size)” in IMediaStream interface - will it work when I connect to onStreamCreate event and addAudioData some aac+v2 bytes with “intro sound”??
That’s all for now, thanks in advance for help.