Re: [hybi] More feedback on WebSockets

[Greg Wilkins <[e-mail]>]

Ian Hickson wrote:
> On Tue, 27 Oct 2009, Greg Wilkins wrote:
>> Ian Hickson wrote:
>>> I disagree that the current handshake isn't like HTTP enough, though. 
>>> The request is fully HTTP-compliant. What value would there be in 
>>> relaxing the rules on what WebSocket clients should send in the 
>>> request? I don't understand the real-world case you are concerned 
>>> about.
>> What about a load balancer in front of the server that inserts a cookie 
>> or X-Forwarded-For header into all HTTP requests that it forwards.
>>
>> This will probably be harmless with regards to the subsequent WS 
>> connection, but it will break the handshake so there will not be a 
>> subsequent WS connection.
> 
> How do such load balancers handle pipelining?
> 
> If they support them, then that means they are almost certainly 
> incompatible with WebSocket, as far as I can tell, and we would _want_ the 
> connection to fail.

They all are different.  But many just look at the first request on
a connection and then just treat the rest of the connection with
packet forwarding.   So that type should work.

Others look at every request and will probably not work.

regards