Brian,
I Hope that this is not a longer answer than you wanted, but here it goes:
>From a generic sense, keepalive usually refers to a mechanism that ensures that an
>established "connection" or "session" is still alive. It is used at different levels
>in the OSI stack.
Here are some examples:
A link pulse ensures that a 10BaseT Physical connection is still up.
Hellos in protocols like OSPF and EIGRP help ensure that neighbors are still reachable.
Novell servers send NCP Watchdog. This ensures that a user connected to the server is
still alive, if not it will terminate the session and free the resources.
SPX has a keepalive timer in which packets are consistently sent accross the
connection, even if no data is being sent.
TCP has optional keepalive. This allows TCP to close connections for which the partner
"went away". This allows the resources used by the connection to be freed.
NetBIOS sessions employ keepalives as well.
I suppose that there are many, many, more examples.
I guess you summarize as follows: Keepalives and keepalive timers (often knwon by
different names) help determine how long a period of inactivty should be tolerated
before: Clearing connections/sessions or determining a link has gone down. This allows
resources to clear, routing tables to converge faster and protocols to operate more
efficiently.
Tom
___________________________________
UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
FAQ, list archives, and subscription info: http://www.groupstudy.com
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
