From owner-freebsd-bugs  Tue Aug 22 00:13:56 1995
Return-Path: bugs-owner
Received: (from majordom@localhost)
          by freefall.FreeBSD.org (8.6.11/8.6.6) id AAA09209
          for bugs-outgoing; Tue, 22 Aug 1995 00:13:56 -0700
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34])
          by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id AAA09193
          for <freebsd-bugs@freefall.FreeBSD.org>; Tue, 22 Aug 1995 00:13:50 -0700
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id RAA27695; Tue, 22 Aug 1995 17:07:19 +1000
Date: Tue, 22 Aug 1995 17:07:19 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199508220707.RAA27695@godzilla.zeta.org.au>
To: bde@zeta.org.au, freebsd-bugs@freefall.FreeBSD.org,
        ponds!rivers@dg-rtp.dg.com, ponds!rivers@zeta.org.au
Subject: Re: Problem with sio at high baud on 2.0.5?
Sender: bugs-owner@FreeBSD.org
Precedence: bulk

> Then, I started pinging the remote host.  After about 30 seconds, all
>modem traffics stops... then, after about a minute, the ping (which
>of course, gets no ICMP packets back) at this point begins issueing a message
>about "No buffer space available."

There was another report yesterday about "No buffer space available"
together with silo overflows.  In your case this is explained by
something retrying the send while the low level remained wedged.

>4 sio lines
>  LINE RAW CAN OUT  HWT LWT     COL STATE  SESS  PGID DISC
>  cuaa0 114   0   0 2052 256  -22912 OCl  582ea0  4991 term
>  cuaa1  0   0  87 2052 256    6382 OCB  5325c0  7627 slip
                                       ^

I assumbe the B state never goes away.  Is this correct?  It means that
either the driver neglects to clear it or output really is wedged.  Try
using a timeout to clear the wait:

	comcontrol /dev/cuaa1 drainwait 1
	kill pid-of-slip

Bruce