SignalR negotiates SSE, but also uses long polling?

Question

I am developing a SignalR 2.0.1 application on a W7 machine, with Chrome, .NET 4.5.1 and IIS 7.5.

The app works well, but digging in the activity I see something weird. First, it negotiates SSE as transport:

[12:54:37 GMT+0000 (GMT Standard Time)] SignalR: Negotiating with 'updates/negotiate?user=true&clientProtocol=1.3'. 
[12:54:37 GMT+0000 (GMT Standard Time)] SignalR: Attempting to connect to SSE endpoint '<url>/updates/connect?transport=serv…bLFxY6YS1OOyLL%2FtuAUH63PQEu2SFoXeYaS9X0yZQu%2BWUJ1TXM%2FG&user=true&tid=3'. 
[12:54:37 GMT+0000 (GMT Standard Time)] SignalR: EventSource connected. 
[12:54:37 GMT+0000 (GMT Standard Time)] SignalR: Now monitoring keep alive with a warning timeout of 13333.333333333332 and a connection lost timeout of 20000. 

And in the Network tab, in the Chrome Developer tools, I can see a request with "text/event-stream" content-type, that despite of looking finished, grows in time and content.

But, I can see also a lot of what seems to be "long polling" connections:

XHR finished loading:  http://localhost:6725/738305d7ff914c0b8abf0d68490625af/arterySignalR/poll?transport=longPolling&connectionToken=<token>&requestUrl=<url>&browserName=Chrome&tid=4&_=1389012957485

I can see how those connections starts, and take a long time to complete.

Why is that happening? At the beginning I thought it was a "keep alive" but why would a keep alive use long polling?

Cheers.

Answer 1

Artery is the code name for browser link. It's runs while you're in VS and you can turn it off if need be.