Bob Doolittle schrieb:
Joerg Barfurth wrote:
Stuart schrieb:
Basically I want to detect if somebody runs off with a sun ray. :-)
You can't reliably detect that across an arbitrary TCP/IP network
connection.
I presume you mean it can't be detected if it's plugged in again
somewhere else? As you say below, if it's not plugged back in it should
be detectable...
You can't detect where the connection is broken (it might be the switch
in the server room) and you can't detect if apparent loss of connection
is due to intermediaries, from which the network might recover
eventually on its own by rerouting.
It is a judgement call after what timeout you declare a connection dead.
In SRSS we have the (very long) 8 minute TCP timeout. And in 4.2 we'll
have a very short (~10 sec) timeout, if a non-responsive connection
blocks a new connection for the same token (under certain circumstances).
What happens if the DTU is plugged into a separate IP subnet and gets a
new IP address assigned? Does this simulate a disconnect and then
reconnect?
Yes.
When we get a new connection for an existing terminal (and, in 4.2, if
the terminal has not failed client authentication) then we internally
disconnect the old connection. This mechanism does not check whether the
connections use the same IP address (even a simple DTU reset may result
in a different IP address, depending on the DHCP setup).
- Jörg
--
Joerg Barfurth phone: +49 40 23646662 / x66662
Software Engineer mailto:[email protected]
Desktop Technology http://reserv.ireland/twiki/bin/view/Argus/
Thin Client Software http://www.sun.com/software/sunray/
Sun Microsystems GmbH http://www.sun.com/software/javadesktopsystem/
_______________________________________________
SunRay-Users mailing list
[email protected]
http://www.filibeto.org/mailman/listinfo/sunray-users
