From mboxrd@z Thu Jan 1 00:00:00 1970 MIME-Version: 1.0 In-Reply-To: References: <80191640-DB1A-4681-9289-8C17E0ED21F9@gmail.com> Date: Wed, 1 Sep 2010 13:44:46 -0400 Message-ID: From: John Floren To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [9fans] how to lock cpu console Topicbox-Message-UUID: 4cbe599e-ead6-11e9-9d60-3106f5b1d025 On Wed, Sep 1, 2010 at 1:22 PM, erik quanstrom wrot= e: > On Wed Sep =C2=A01 12:58:54 EDT 2010, benavento@gmail.com wrote: >> you right, I thought conslock was rob's lock program >> >> http://plan9.bell-labs.com/sources/patch/sorry/robs-bits/ > > i hate doing this, but that depends on rio, too. =C2=A0the open of > /dev/screen -> error() -> exits("fatal error"); > > - erik > > Thus, screenlock is like rob's lock program... rio is such a minor thing to run on today's massive machines, I'm not sure I really see the problem in starting it on your cpu server anyway. I frequently set them up to launch into rio because: 1. It's easier to fix things when I can cat /dev/kprint in a window rather than have it constantly interrupting me 2. I like to be able to interrupt programs 3. It's nice to run more than one thing at once, have a graphical editor, e= tc. 4. Full-screen stats is pretty Of course, none of these reasons matter to you, since you don't run rio on your servers AND you don't think there's any reason to lock them (I agree!), I'm just pointing out that graphical lockers and rio in general are far from useless on a cpu server. John --=20 "With MPI, familiarity breeds contempt. Contempt and nausea. Contempt, nausea, and fear. Contempt, nausea, fear, and .." -- Ron Minnich