[TriLUG] iptables & FUD

Aaron Joyner aaron at joyner.ws
Mon Apr 29 12:56:57 EDT 2013


Careful not to go too far into Grumpy Old Man mode.  :)  Personally, I do
like these sorts of changes.  The new tools often provide a lot simpler /
more intuitive mapping of what's going on in the kernel.  In some cases,
they provide functionality that's simply *not* available with the old
tools.  Multiple IP addresses per interface is a complete *hack* with
ifconfig (eth0:1?  really?).  VLANs get even worse (which physical
interface is vlan10 bound to?  `ifconfig vlan10` has no idea, but `ip link
show` does).  To the best of my knowledge, you simply can't do policy
routing[1] with the plain 'route' command.

I use block and line highlighting in vim on a daily basis, but I still know
how to :.,+3s/foo/bar/ when the time calls for it.  :)

My point is that knowing one set is powerful.  Almost all of these tools
(not quite all, I did mention Emacs) are excellent examples of the Unix
Philosophy[2].  With any of these choices, you reap the benefits of that
way of thinking.  Knowing more than one way to do things, and being able to
know when to choose one way over the other (new features or mental models
vs maximum portability and user familiarity) is one of those things that
separates the grasshopper from the master.

Aaron S. Joyner
Perpetual Grasshopper


1 - Note that the "new" tools aren't all that new, the following link is
from > 9 years ago:
http://www.trilug.org/pipermail/trilug/Week-of-Mon-20040329/025177.html
2 - IMHO, the canonical description of the Unix Philospohy is captured in
Eric S. Raymond's book, The Art of Unix Programming.  It's available free
online, or in dead tree edition for your book shelf.  Everyone who works
with Unix should read it at least once.  The link that's most relevant to
this discussion: http://www.catb.org/esr/writings/taoup/html/ch01s06.html


On Mon, Apr 29, 2013 at 12:01 PM, William Sutton <william at trilug.org> wrote:

> Oh, I agree with you... I just want someone from the camp that likes to
> make these sorts of changes to defend why wholesale rearrangment of the
> standard UNIX tools and subsystems is a good idea....because I don't think
> it is, but clearly someone with more authority over what does (or doesn't)
> go into some generic Linux system seems to think so.
>
> William Sutton
>
>
> On Mon, 29 Apr 2013, David Both wrote:
>
>  That is the precise point; "old" does not mean obsolete, especially when
>> talking about Unix and GNU utilities that were designed with the Unix
>> philosophy in mind.
>>
>> If you don't know what the "Unix/Linux Philosophy" means, grasshopper,
>> try these links.
>>
>> http://read.pudn.com/**downloads63/ebook/222048/**
>> Linux%20and%20the%20Unix%**20Philosophy.pdf<http://read.pudn.com/downloads63/ebook/222048/Linux%20and%20the%20Unix%20Philosophy.pdf>
>> http://en.wikipedia.org/wiki/**Unix_philosophy<http://en.wikipedia.org/wiki/Unix_philosophy>
>>
>>
>>
>> On 04/29/2013 11:42 AM, William Sutton wrote:
>>
>>> vim +1
>>>
>>> that said, I'm sick of people changing (and/or deprecating) perfectly
>>> functional software just because they don't think it's "pretty" enough, or
>>> is organized the way they would do it.  See, in addition to the below
>>> example, the udev-197 changes, which completely hosed my PC for a week.
>>>
>>> Yes, I get that sometimes things are obsolete, or the maintainer closed
>>> the source, or Oracle bought yet another tool, or (etc., etc., etc.).  But
>>> some things I just don't get.
>>>
>>> William Sutton
>>>
>>> On Mon, 29 Apr 2013, Aaron Joyner wrote:
>>>
>>>  If you aspire to a career in system administration, or simply want to
>>>> work
>>>> with embedded systems, it is as important to know how to do things with
>>>> 'netstat' and 'route' as it is to know which sexy features of 'vim'
>>>> aren't
>>>> supported in vanilla 'vi'.  When you deal with commercial unixes
>>>> (unices?
>>>> eg. Solaris, HPUX or AIX), you're likely to be dealing with the
>>>> "classic"
>>>> tools such as netstat, you certainly won't have the new-hotness of
>>>> iproute2
>>>> (iprule is *right* *out*).  On an embedded linux distros (Montavista,
>>>> anything with a busybox core, probably even the Rasberry Pi?), the first
>>>> thing to go when space is at a premium are duplicate tools.  I have yet
>>>> to
>>>> find a developer that's favored dropped the classic tools over the
>>>> new-shiny tools, although I suppose some day that's coming. Even when we
>>>> cross that Rubicon, they're still likely to include a package
>>>> equivalent to
>>>> Debian's vim-tiny rather than full-blown vim.  Typically, you can forget
>>>> emacs (and as a general rule, you should).
>>>>
>>>> Aaron S. Joyner
>>>> (starts vi/emacs flame war on mailing list, goes to lunch)
>>>>
>>>>
>>>> On Mon, Apr 29, 2013 at 9:51 AM, John Vaughters <jvaughters04 at yahoo.com
>>>> >wrote:
>>>>
>>>>  Also agreed, I get frustrated with some of the newer tools that greatly
>>>>> increase your typing. To the point that I start to wonder if I am on
>>>>> the
>>>>> path of Grandpahood, where we curse all new and stick to all old
>>>>> schooling
>>>>> those whipersnapers at every chance, only to one day to die a good John
>>>>> Henry death by the true new technologies that you never saw coming.
>>>>>
>>>>> The Cycle of Life!
>>>>>
>>>>>
>>>>> ______________________________**__
>>>>> From: David Both <dboth at millennium-technology.**com<dboth at millennium-technology.com>
>>>>> >
>>>>> To: trilug at trilug.org
>>>>> Sent: Monday, April 29, 2013 9:43 AM
>>>>> Subject: Re: [TriLUG] iptables & FUD
>>>>>
>>>>>
>>>>> I find that many of the very oldest solutions are still the most
>>>>> elegant.
>>>>> They
>>>>> always have the advantage of simplicity, being written to work well
>>>>> with
>>>>> limited
>>>>> resources of all types, conforming (for the most part) to the
>>>>> Unix/Linux
>>>>> philosophy, and having been thoroughly debugged over many years.
>>>>> --
>>>>> This message was sent to: Aaron S. Joyner <aaron at joyner.ws>
>>>>> To unsubscribe, send a blank message to trilug-leave at trilug.org from
>>>>> that
>>>>> address.
>>>>> TriLUG mailing list : http://www.trilug.org/mailman/**listinfo/trilug<http://www.trilug.org/mailman/listinfo/trilug>
>>>>> Unsubscribe or edit options on the web  :
>>>>> http://www.trilug.org/mailman/**options/trilug/aaron%40joyner.**ws<http://www.trilug.org/mailman/options/trilug/aaron%40joyner.ws>
>>>>> TriLUG FAQ          :
>>>>> http://www.trilug.org/wiki/**Frequently_Asked_Questions<http://www.trilug.org/wiki/Frequently_Asked_Questions>
>>>>>
>>>>>  --
>>>> This message was sent to: William <william at trilug.org>
>>>> To unsubscribe, send a blank message to trilug-leave at trilug.org from
>>>> that address.
>>>> TriLUG mailing list : http://www.trilug.org/mailman/**listinfo/trilug<http://www.trilug.org/mailman/listinfo/trilug>
>>>> Unsubscribe or edit options on the web    :
>>>> http://www.trilug.org/mailman/**options/trilug/william%**40trilug.org<http://www.trilug.org/mailman/options/trilug/william%40trilug.org>
>>>> TriLUG FAQ          : http://www.trilug.org/wiki/**
>>>> Frequently_Asked_Questions<http://www.trilug.org/wiki/Frequently_Asked_Questions>
>>>>
>>>>
>>>> --
>>>>
>>>>
>>>> ***********************************************************
>>>> David P. Both, RHCE
>>>> Millennium Technology Consulting LLC
>>>> 919-389-8678
>>>>
>>>> dboth at millennium-technology.**com <dboth at millennium-technology.com>
>>>>
>>>> www.millennium-technology.com
>>>> www.databook.bz - Home of the DataBook for Linux
>>>> DataBook is a Registered Trademark of David Both
>>>>
>>>>  --
>> This message was sent to: William <william at trilug.org>
>> To unsubscribe, send a blank message to trilug-leave at trilug.org from
>> that address.
>> TriLUG mailing list : http://www.trilug.org/mailman/**listinfo/trilug<http://www.trilug.org/mailman/listinfo/trilug>
>> Unsubscribe or edit options on the web  : http://www.trilug.org/mailman/*
>> *options/trilug/william%**40trilug.org<http://www.trilug.org/mailman/options/trilug/william%40trilug.org>
>> TriLUG FAQ          : http://www.trilug.org/wiki/**
>> Frequently_Asked_Questions<http://www.trilug.org/wiki/Frequently_Asked_Questions>
>>
>>  --
> This message was sent to: Aaron S. Joyner <aaron at joyner.ws>
> To unsubscribe, send a blank message to trilug-leave at trilug.org from that
> address.
> TriLUG mailing list : http://www.trilug.org/mailman/**listinfo/trilug<http://www.trilug.org/mailman/listinfo/trilug>
> Unsubscribe or edit options on the web  : http://www.trilug.org/mailman/**
> options/trilug/aaron%40joyner.**ws<http://www.trilug.org/mailman/options/trilug/aaron%40joyner.ws>
> TriLUG FAQ          : http://www.trilug.org/wiki/**
> Frequently_Asked_Questions<http://www.trilug.org/wiki/Frequently_Asked_Questions>
>



More information about the TriLUG mailing list