From mboxrd@z Thu Jan  1 00:00:00 1970
From: rsc@plan9.bell-labs.com
Message-Id: <200006151914.PAA06832@smtp1.fas.harvard.edu>
Date: Thu, 15 Jun 2000 15:14:29 -0400
To: 9fans@cse.psu.edu
Subject: Re: [9fans] UNIX to Plan9 Command Translation?
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Topicbox-Message-UUID: bf50098e-eac8-11e9-9e20-41e7f4b1d025

Brian Kernighan wrote for the second edition,
``Plan 9 is not UNIX.  If you think of it as UNIX,
you'll often be frustrated because something doesn't
exist or works differently.  If you think of it as
Plan 9, however, you'll find that most of it works
very smoothly, and that there are some truly neat
ideas that make things much cleaner than you
have seen before''.

That quoted, I sometimes miss df and uptime too.

    df(1)                   ?

There's no way to do this in the file protocol,
so in general, you can't.  The last paragraph
of the "The File Server" section in the
Plan 9 from Bell Labs paper explains one reason why.

A shell script a bit large to post here works
on kfs partitions.

hget http://plan9.bell-labs.com/~rsc/plan9/df

    more(1)                 p (I want less(1), actually.)

Experiment with scroll mode and noscroll mode
in rio's button 2 menu.

    shutdown(1m)            disk/kfscmd halt (and?)

And Ctl-Alt-Del or ^T^Tr.  If you're booting
from a network file server (and thus have no
kfs local disk), you can just turn the machine off.
It takes a little getting used to.

    id(1)                   ?

cat /dev/user
echo $user
grep $user /adm/users

id(1) doesn't really make too much sense, since
who you are depends on which file server you are
speaking with.  I am the plan9.bell-labs.com rsc
on some 9P connections, the eecs.harvard.edu rsc on
other 9P connections, and just none on some u9fs
over SSH connections.

    find(1)                 ? (!)

du -a | awk '{print $2}' is usually a good start.

    apropos(1)              ?

man(1) is usually enough that this hasn't been done.

    uptime(1)               ?

cat /dev/time (see cons(3))/

A shell script a bit too ugly to post here is at
hget http://plan9.bell-labs.com/~rsc/plan9/uptime,

    netstat(1m) [-r flag]   ?

cat /dev/iproute (see ip(3))

    last(1)                 ?

This doesn't make a whole lot of sense even in
UNIX.  Since we don't have ttys, it makes even
less sense here -- when do you log and when do
you not?  What's the difference between someone
who is logged in and a cron job running as that
person?  How can you tell?

    vmstat(1m)              ?

cat /dev/swap
stats -mw

gets you some of that.

    iostat(1m)              ?
    nfsstat(1m)             ? (for 9fs)

Poke around in '#z' and have a look at
/sys/src/9/port/devmntstats.c

    snoop(1m)		    ?
    tcpdump(1)              ?

snoopy(8)

    ufs{dump|restore}(1m)   ?
    pkginfo(1m)             ?

mkfs(8) and wrap(8).

Russ