> It's a common socket programming problem. Usually, one isn't
> too worried about
> it because processes tend to stay alive, keep the socket open
> and bound, and
> listen to subsequent requests, rather than being killed and
> having to restart,
> open and bind a socket. Perhaps some kind of mech
On Tue, 21 Aug 2001 09:32:42 -0400 "Jeff Johnson" <[EMAIL PROTECTED]> wrote:
>
>Hey Paul, great suggestion, I think you win the prize. Here's a bunch
>of netstats taken roughly a second apart. The TIME_WAIT socket was
>opened to send a single GIF to my browser. I closed the browser to
>make sur
> What does netstat -i show for the socket bound to port 8086?
> I would expect
> something like one of the WAIT states, meaning that the
> socket is hanging
> around waiting for all data to be exchanged and for both ends
> of the connection
> to "agree" on the closure of the connection.
>
> Inter
"Jeff Johnson" <[EMAIL PROTECTED]> wrote:
>
>Yep, this is about BSD not releasing the 8086 socket at shutdown and
>subsequent calls to appserver hanging when they try to bind to it.
>
>If you have anything you'd like me to test, just let me know :)
What does netstat -i show for the socket bound t
