Re: [rtcweb] draft-ibc-rtcweb-sip-websocket -- WebSocket Transport for Session Initiation Protocol (SIP)

[Christer Holmberg <[e-mail]>]

Hi,

Thanks for putting the draft together.

I agree with Markus, and others, that WebSocket is a feasible solution in the case of using a JS SIP app.

>> 1) in Section 3, its mentioned that there is no real 
>> benefit of using SIP over Websockets. If this is the case, 
>> why are you proposing integrating SIP with Websockets?
> 
> As Saul pointed out, the text says:
> 
>    There is no benefit on using SIP over WebSocket  transport between
>    two SIP nodes when none of them runs within a web browser.
> 
> JavaScript can open a WebSocket connection from a web browser 
> to a server, but not an arbitrary TCP or UDP connection.

I have a comment, related to this.

The text says:

   "Even more, the WebSocket protocol is not symmetric since
   just a WebSocket client can open a connection to a WebSocket server
   (a WebSocket client does not listen for incoming connections)."

But, this is of course also true when you DO use a browser, but it's not really addressed in the draft. You need to describe that it is assumed that the WebSocket connection has been established at the point when the browser is supposed to be able to receive incoming SIP calls.

For example, the WebSocket could be established once the JS app is downloaded, or when the JS SIP client registers (similar to how SIP Outbound works).


>> 3) In Section 4, you mentioned that since Websocket is a 
>> reliable transport protocol, the websocket sub-protocol 
>> defined for SIP also becomes reliable. I don't agree with 
>> this - The websocket connection is actually initiated by the 
>> web browser client towards the web server and this connection 
>> needs to be kept alive through some keep alive mechanism. 
>> Otherwise the connection may close. I am not sure of reliability.
>> 
> "Reliability" does not mean that. "Reliability" means data 
> delivery acknowledgment in the transport. This is true in TCP 
> (so also in WebSocket which works on top of TCP) and false in 
> UDP (which is not a reliable transport protocol).
> 
> The keep-alive mechanism is needed for cases in which there 
> is a NAT router in the path. Those routers usually close a 
> TCP connection after some minutes of inactivity. This happens 
> in any TCP connection from client to server through a NAT box.

Since WebSocket provides a keep-alive function, you can of course use that. I guess it could still be mentioned that a SIP level mechanism exists.

Regards,

Christer