tcp_server_async bug fixed.

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
Anonymous
Not applicable

Hi All

i found the TCP stack of NetX_duo has a bug based on SDK 2.4.0.

The demo App is snip.tcp_server_async

The wiced modules was running as server, and other device connect to server will be successfully for the first time.

If the client disconnected from server, the callback function "client_disconnected_callback" will be called, the server will start to listen on socket again to wait for the next connection from client, unfortunately the listen action will return error which will cause the client can't connect to server again after the 1st disconnection from server.

So we should call wiced_tcp_delete(tcp_socket) first, then redo the wiced_tcp_create_socket, wiced_tcp_register_callbacks, and wiced_tcp_listen on new created socket again.  The server can be connected and disconnected successfully every time by client.

i am not sure if the netx_duo stack has bug!

Hope you enjoy it.

Regards

Harry

2 Replies
GregG_16
Employee
Employee
50 sign-ins 25 sign-ins 25 comments on KBA

Harry, 

Thank you for the fix and the details. 

0 Likes

Is there a fix for this issue available, or do we still have to use the work-around outlined in the original post?  I am finding limitations in the speed with which WICED can build and tear down TCP/IP connections.  For Modbus/TCP, which specifies a separate connection for each transaction, this becomes a real problem.

0 Likes