It will need about 18 mins to stop connection(disconnecting to disconnected) when it lost network(Like disble WiFi insterface). And it wil cause Linux Segmentation Fault when I try to call hub_connection::stop() while connection state is disconnecting.

[G-MoYang]

Is there an existing issue for this?

• I have searched the existing issues

Describe the bug

Envoriment: Raspberry pi， Raspbian GNU/Linux 11

Steps:

1. Make connection with server.
2. Then dsable the network insterface to make it lost network.

Two issue:

1. It will need 18 mins to stop connection(disconnecting to disconected). How I can change closing handshake timed out.
   

2. It will cause Linux Segmentation Fault I try to call hub_connection::stop() while connection state is disconnecting.
   

Expected Behavior

No response

Steps To Reproduce

No response

Exceptions (if any)

No response

.NET Version

No response

Anything else?

No response

[G-MoYang]

The branch is https://github.com/aspnet/SignalR-Client-Cpp

[BrennanConroy]

Looks like you're using an older version of the client. Try updating and seeing if the issue persists.

Hi @G-MoYang. We have added the "Needs: Author Feedback" label to this issue, which indicates that we have an open question for you before we can take further action. This issue will be closed automatically in 7 days if we do not hear back from you by then - please feel free to re-open it if you come back to this issue after that time.

[G-MoYang]

Okay, let me try to update it and feedback

[G-MoYang]

@BrennanConroy I have tried to upgrade to last version of client. But it still need 18mins to disconnected and it wil cause Linux Segmentation Fault when I try to call hub_connection::stop() while connection state is disconnecting.

How i can stop "websocket transport" quickly after lost network connetion?

[BrennanConroy]

18 minutes is roughly how long a TCP connection takes to timeout on Linux: https://stackoverflow.com/questions/13085676/tcp-socket-no-connection-timeout

This implies that the websocket isn't closing through normal means, which would normally take 5 seconds if the server is gone.

Did you provide a custom websocket implementation? Can you turn on trace logging?
It looks like you have a backtrace handler, can you get the stack trace of the seg fault?

[G-MoYang]

I am using the cpprestsdk[websockets] in vcpkg's ports. I try to enable the trace logging in websocketpp by:
m_alog->set_channels(log::alevel::all);
m_elog->set_channels(log::alevel::all);

And for the crashing issue, it is the replicating steps:

1. Create a local variable hub_connection connection by call hub_connection_builder::create(url) in a function.
2. When it connection is disconnecting, try to quit the function.
3. Then it will crash in some destructor fucntion
4. I enable the backtrace of SignalR library, it assert here:
   hub_connection_impl.cpp->stop->assert(connection->get_connection_state() == connection_state::disconnected);

[G-MoYang]

Add trace logging:

it seem it will timeout in 10s, but not sure it not callback to m_underlying_client.close().then

[G-MoYang]