mina-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Trustin Lee" <trus...@gmail.com>
Subject Re: Additional questions regarding MINA 0.9.5
Date Tue, 05 Sep 2006 01:13:04 GMT
On 9/4/06, Kok Hoor (Gmail) <kokhoor@gmail.com> wrote:
> Hi,
> There is this slight issue that I face with Mina. It is a minor issue only
> though.
> I am using Windows XP as the development and testing environment, and I
> have
> three separate instances of Mina running, one is the Server simulator,
> another is the Proxy and the last is the Client simulator. When I press
> Ctrl-C in the Proxy, and the Proxy terminates, the Client, which uses an
> IOConnector, and is still sending messages to the Proxy will receive an
> exception from the Proxy:
>         java.io.IOException: An existing connection was forcibly closed by
> the remote host
> My exceptionCaught function will close the Session. The code for my
> function
> is below:
>     public void exceptionCaught( IoSession session, Throwable cause )
>     {
>         SessionLog.info( session, "SxCommServerSessionHandler::Exception:"
> +
> cause);
>         cause.printStackTrace();
>         // close the connection on exceptional situation
>         session.close();
>     }
> However, I still need to wait for around 10 - 30 seconds before my client
> application exits. I have already inserted 'session.isClosing()' calls
> before doing writes to prevent future writes from blocking the close. Is
> it
> due to NIO? Or is there anything I can do to explicitly to increase my
> client shutdown time? It is funny because using conventional Java sockets,
> I
> do not this problem.

Well, usually, all servers should provide a way to shut itself down
gracefully such as:

1) unbind the port
2) close all connections
3) exit

It is a little bit weird that you are getting the exception even if you
terminated the server with CTRL+C.  It is usually thrown when you killed a
process with -9 option.

what we call human nature is actually human habit
PGP key fingerprints:
* E167 E6AF E73A CBCE EE41  4A29 544D DE48 FE95 4E7E
* B693 628E 6047 4F8F CFA4  455E 1C62 A7DC 0255 ECA6

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message