From mboxrd@z Thu Jan  1 00:00:00 1970
Message-ID: <14515.98067.qm@web83904.mail.sp1.yahoo.com>
Date: Thu, 25 Feb 2010 07:22:54 -0800
From: "Brian L. Stuart" <blstuart@bellsouth.net>
To: 9fans@9fans.net
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Subject: [9fans] Factotum discrepancy
Topicbox-Message-UUID: db42db14-ead5-11e9-9d60-3106f5b1d025

There appears to be a disagreement between the=0Afactotum man page and its =
actual behavior regarding=0Athe -a option.=A0 In the man page, the wording:=
=0A=0A-a=A0=A0=A0supplies the address of the authentication server to=0A=A0=
 =A0=A0=A0use.=A0 Without this option, it will attempt to find an=0A=A0 =A0=
=A0=A0authentication server by querying the connection=0A=A0 =A0=A0=A0serve=
r, the file <mtpt>/ndb, and finally the network=0A=A0 =A0=A0=A0database in =
/lib/ndb.=0A=0Amakes it sound like the -a option takes precedence=0Aover cs=
 and ndb.=A0 But the code for _autdial() in=0Autil.c looks for cs first and=
 if it's there, the=0A-a option is ignored.=A0 (In p9p, -a seems to be igno=
red=0Aaltogether.)=0A=0AWhich way should it be?=A0 I'd prefer if -a took pr=
ecedence,=0Abut if the existing behavior is really what's wanted,=0Athen I =
suggest we change the man page so it's more clear=0Athat's what's happening=
.=0A=0ABLS=0A=0A