Question

As I know, The OnReconnected event handler in a SignalR Hub can execute directly after OnConnected but not after OnDisconnected for a given client. (source: http://www.asp.net/signalr/overview/guide-to-the-api/handling-connection-lifetime-events)

So, if onReconnected will never be happened after onDisconnected and context.connectionId will remain the same, why official example checks context.connectionId in user connections and add it if not exists.

Link: http://www.asp.net/signalr/overview/guide-to-the-api/mapping-users-to-connections

Answer 1

  public override Task OnDisconnected(bool stopCalled)
        {
            string name = Context.User.Identity.Name;

            _connections.Remove(name, Context.ConnectionId);

            return base.OnDisconnected(stopCalled);
        }

如果正常断开连接，则stopCalled将为true。否则，它将是假的（超时等），但这并不意味着信号器肯定会断开连接。

如果SignalR位于配置了scaleout的负载均衡器后面，那么客户端可能仍然连接到另一个SignalR服务器。

即使客户端未断开连接，也可以触发OnDisconnected(false)。然后删除Context.ConnectionId。

但是如果客户端仍然连接，那么将重置连接。因此，您应该检查的时间可能是您使用OnDisconnected(false)删除了此连接，这不是实际断开连接。

Why are we checking context.connectionId when a recconect in signalR occurs?

1 个答案: