>That doesn't sound plausible to me.  When the queues back up, log
>messages usually get dropped, rather than blocking the application.

Yes, kernel messages get dropped, but if syslogd makes a write(2) to
/dev/console, is it smart enough to time this out and or fail gracefully
if the console is "stuck"?

When I was trying to fix the serial ports a long time ago, one of the
behaviors I saw was that a write(2) to the serial would hang, un-killably.
The system had to be rebooted to get the proc doing the write(2) to
go away.

-dgl-