🚀 go-pugleaf

In <3lc25d$kt1@cpcug.org>, kbennett@cpcug.org (Keith Bennett) writes:
>1) Is this behavior correct according to the TCP/IP standard (or
>   convention if there isn't a standard)?

yes.  tcp/ip is a packet-switched (rather than a
connection-switched) protocol simply because the
designers didn't want TCP "connections" to drop when
the network was flakey.  just because you haven't
received any messaged for several hours/days/weeks,
doesn't mean that a TCP "connection" has been severed.

>2) Is there a way to get the recv() to detect the shutdown and
>   return an error code?

use the KEEPALIVE option.  with KEEPALIVE, the stack
will send a packet every few minutes or hours.  if it
doesn't receive an acknowledgement, the next recv() will
fail.

>   A slightly different solution would still have a separate thread
>   checking the other machines, but instead of pinging have a
>   separate special purpose socket with periodic keepalive
>   messages.  It is our understanding that sockets are more
>   reliable than ping (tcp vs. udp) and that sending messages
>   across an open socket would require less overhead than ping.
>   (Is that correct?)

no, it's totally bogus.  how do you think PING
works anyway?  (a hint: it doesn't use either TCP
or UDP.)
__
\/ Jake Coughlin, [jakec@vnet.net (preferred), coughlin@wg.com]
"I suspect the INI files. I almost always suspect the INI files. I hate
those things. They are full of deep dark secrets, unfathomable to most
of the great unwashed!" -- Darcy B. Hager (Darcy.Hager@UAlberta.CA)