From dbslinux at bellsouth.net  Mon Oct  8 13:41:22 2001
From: dbslinux at bellsouth.net (DBSLinux)
Date: Mon, 8 Oct 2001 16:41:22 -0400
Subject: [buug] DNS problem?
Message-ID: <002301c15039$979273c0$0a00a8c0@dbsinc>

I am trying to setup a linux server as a router. Right now I can use lynx to browse the web from the linux server. From the WIN98 workstations I can ping the linux server and can ping the dns server but can not ping anything that is using a name. I can not get any of the win98 pcs to browse the web. Can anyone give some ideas where to start looking for a solution to this problem. 

Regards!!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://buug.org/pipermail/buug/attachments/20011008/4b11e492/attachment.html>

From gnuman at earthlink.net  Mon Oct  8 14:20:33 2001
From: gnuman at earthlink.net (Blake Golliher)
Date: Mon, 8 Oct 2001 14:20:33 -0700
Subject: [buug] DNS problem?
In-Reply-To: <002301c15039$979273c0$0a00a8c0@dbsinc>
References: <002301c15039$979273c0$0a00a8c0@dbsinc>
Message-ID: <01100814203300.19613@cuda>

On Monday 08 October 2001 13:41, DBSLinux wrote:
> I am trying to setup a linux server as a router. Right now I can use lynx
> to browse the web from the linux server. From the WIN98 workstations I can
> ping the linux server and can ping the dns server but can not ping anything
> that is using a name. I can not get any of the win98 pcs to browse the web.
> Can anyone give some ideas where to start looking for a solution to this
> problem.
>
> Regards!!

----------------------------------------
Content-Type: text/html; charset="iso-8859-1"; name="Attachment: 1"
Content-Transfer-Encoding: quoted-printable
Content-Description: 
----------------------------------------


Sounds like your not cacheing right.

check out http://www.linuxdoc.org/HOWTO/DNS-HOWTO.html for some info on that, 
particularly the caching info.  It sounds like you are using your ISP's DNS 
servers, on your linux server, and your clients aren't able to, because they 
can't resolve ns1.isp.net to the ip of it.


Blake ~=)



From alex at myzona.net  Mon Oct  8 14:22:33 2001
From: alex at myzona.net (Alex M)
Date: Mon, 8 Oct 2001 14:22:33 -0700
Subject: [buug] DNS problem?
References: <002301c15039$979273c0$0a00a8c0@dbsinc>
Message-ID: <000d01c1503f$5ad11c60$9601a8c0@parkson>

Anything could be a problem, you have not provided us with enough
information and network configuration you have.
Frankly I dont  see how can DNS be related here. Setup a NAT server on your
linux machine in order to have win98 machines access the internet.

----- Original Message -----
From: "DBSLinux" <dbslinux at bellsouth.net>
To: <buug at weak.org>
Sent: Monday, October 08, 2001 1:41 PM
Subject: [buug] DNS problem?


I am trying to setup a linux server as a router. Right now I can use lynx to
browse the web from the linux server. From the WIN98 workstations I can ping
the linux server and can ping the dns server but can not ping anything that
is using a name. I can not get any of the win98 pcs to browse the web. Can
anyone give some ideas where to start looking for a solution to this
problem.

Regards!!





From cmsclaud at arches.uga.edu  Wed Oct 17 22:18:42 2001
From: cmsclaud at arches.uga.edu (Claude Rubinson)
Date: Thu, 18 Oct 2001 01:18:42 -0400 (EDT)
Subject: [buug] normalization
Message-ID: <Pine.A41.4.10.10110171850590.158064-100000@archa15.cc.uga.edu>

at the last meeting we discussed the advantages and disadvantages of
database normalization.  i just came across some notes that i had written
up last year when i was trying to convince the powers-that-be at my
company that we were in severe need of a database redesign.

(originally, i had just cut-and-pasted the notes into this email message.
then i decided to edit out the company-specific pieces.  as i was editing
out those pieces, i started updating the notes to incorporate what i've
learned since i originally wrote those notes.  in the end, this turned
into a completely new write up.  things sometimes get away from
you.  hopefully, i haven't wasted too much time and someone will get
something of value out of this.)


first things first:

my experience has been primarily with Microsoft's SQL Server.  everything
that i've read suggests that, for the most part, they should generalize to
other high-end relational databases such as Oracle and Sybase.  they'll
generalize less to mid-range relational databases such as PostgreSQL.  
and, of course, they will only generalize to a limited extent to
non-relational databases such as FilePro, Access, and MySQL.  (and i don't
care what people say, MySQL is not a relational database.  it's a fine
product -- i use it myself -- and it has relational features.  but it's
not a relational database.)


benefits of normalization:

A. data integrity

the primary thing that normalization gives you is data integrity.  with a
fully normalized database design, you're not only assured that your
transactions will complete or rollback, you're also assured that inserts,
updates, and deletes -- that is, any type of modification to the data 
-- can't compromise your data.  to my mind, it is this latter attribute
that is the central benefit of relational databases.  there are a variety
of programmatical ways to check that a transaction has fully committed.
but only a fully normalized database with referential integrity can
guarantee the absence of insertion, update, and deletion anomalies.


B. data retrieval

many people will tell you that it's more difficult to build queries
against a fully normalized database.  to a certain extent, that is true.
a fully normalized database will have many tables, each with a handful of
columns.  to retrieve any single piece of data, then, requires joining
together multiple tables.  the key here then is simply to learn SQL -- SQL
was specifically designed to facilitate the retrieval of data from a fully
normalized database.  the problem is breaking out of the 
procedural-oriented programming mentality that most of us are familiar 
with.  indeed, SQL is set oriented which makes it a whole new ballgame.

i've found that with a fully normalized database, i can retrieve any piece
of data with a single query.  and, generally, the queries aren't that
complex.  this isn't to say that i never make use of multiple-statement 
queries; indeed, i often do.  but only under particular circumstances such
as calculations where i'll often need to write one statement per variable
in the calculation.  the important point, however, is that i never need to
run multiple statements to retrieve a given piece of information.


C. performance

anytime you read about normalization, you'll read that normalization
decreases performance.  the idea is that joins between tables are
expensive; since queries of a highly normalized database tend to require
joins between a greater number of tables, performance will suffer.  the
solutions, then, is careful denormalization of the database.

my experience is that this isn't true unless your tables have millions of
records.  a fully normalized database should exhibit greater performance
because (1) your queries and shorter and simpler (as discussed above), (2)
your tables have fewer records since duplicate information has been
eliminated through the normalization process, and (3) indexes are more
efficient because each table only contains a limited amount of
non-redundant information.  this last point can't be overstated as indexes
are a -- if not the -- primary key to performance.  with a properly
normalized database, your queries will almost always hit the clustered
index.


disadvantages of normalization:

A. normalization procedures

despite what i may say sometimes, the proper normalization of a database
is a challenging process.  there are CASE tools which purport to assist in
this process but i've never been particularly impressed with them.  i do
all of my normalization by hand.  the risk of course, is that an
improperly normalized database can cause significant problems.  i once
over-normalized a database such that the query ended up returning false
data.  pretty impressive, i thought.

furthermore, planning is even more important because databases don't lend
themselves to rapid development. the unix philosophy of "build a prototype
early" is a sure recipe for disaster when applied to databases.  i know
this from a very, very painful experience (which i'm still living with
today.  'nuff said).  i now spend a great deal of time researching my
projects before i even begin the design.  


B. database restructuring

proper database design is also important because it can be very difficult
to modify the structure of a database after the fact.  modifying a
properly normalized database is easier than modifying an improperly 
normalized database (primarily because you're more confident in the 
integrity of your data) but modifying a database structure is never
pleasant.

database restructuring is also problematic since the database generally
serves as the foundation of an larger application.  changes to the
database risk breaking the application.  (you can alleviate some of this
risk by implementing abstraction layers between the database and the
application but it's an imperfect solution) 


C. data insertion and maintenance

while a fully normalized database makes it easy to retrieve data, it can
be a serious pain in the ass to insert new records and update existing
records.  i have a handful of tables with inter-table dependencies (i.e.,
they have referential integrity on themselves).  these tables are locked
down so tightly that you can only insert, update, or delete data by
acting on all related records; it's impossible to update individual
records.  this is an extreme example (i've never seen it anywhere else)
but it does occur.  that data maintenance is tightly regulated isn't a bad
thing -- it guarantees the integrity of the data.  but it definitely can
be a pain.


one final thought about relational databases -- they're not always
appropriate.  i've come to believe that relational databases are used far
more often than they need to be.  when people come to me asking about
implementing a database system, more often than not, i end up recommending
against implementing a relational database.  for many applications, text
files are perfectly suitable.  if your needs are limited, the
relationships between your data are not particularly complex, and your
queries are  straightforward, MySQL or FilePro is the way to go.  and if
you find that you're going to need to embed a lot of complex business
rules into your database structure, take a look at an object-oriented
database system.  many of today's complex web applications -- particularly
banking and high-volume community sites such as Slashdot -- would be
better served by an OODBMS.  unlike an RDBMS which segregates the data
from the application, in an OODBMS, the data is the application.
unfortunately, at present, object-oriented database systems tend to be
quite expensive; and i'm not aware of any open source projects.









From cmsclaud at arches.uga.edu  Fri Oct 19 10:44:18 2001
From: cmsclaud at arches.uga.edu (Claude Rubinson)
Date: Fri, 19 Oct 2001 13:44:18 -0400 (EDT)
Subject: [buug] email address?
Message-ID: <Pine.A41.4.10.10110191341240.229194-100000@archa15.cc.uga.edu>

at the end of last night's meeting, there were just two of us left.  i was
one of them.  my memory for names is as bad as my ability for getting
email addresses, so can the other individual shoot me an email?

thanks,

claude





From feedle at feedle.net  Mon Oct 22 20:30:44 2001
From: feedle at feedle.net (C. Sullivan)
Date: Mon, 22 Oct 2001 20:30:44 -0700
Subject: [buug] Re: [Buug-admin] Email List Rental
In-Reply-To: <0EAC4A3C6A28D51192C20090278733053505F3@PDC-WEB2WEB>
References: <0EAC4A3C6A28D51192C20090278733053505F3@PDC-WEB2WEB>
Message-ID: <E15vsH7-0000pa-00@neptune>

On Monday 22 October 2001 16:31, you wrote:
> dear BUUG administrator(s)
>
> I would like to promote a service to members of your user group and invite
> them to an industry event. How can I do that?
>
> Thanks,
>
> Vincent Fallourd
> Alliance Program Manager
> (415) 374-8231
> Acteva
> 633 Battery Street, 2nd Floor
> San Francisco, CA 94111

Please, please... those of you in other UNIX-oriented user groups... please 
tell me that this sort of request is not common... and better yet, please 
tell me that it is never honored...

(Yes, I told the guy to get stiff.  Topic was his, BTW.  Rental?  No.)

-Fedl



From sobolak at myrealbox.com  Tue Oct 23 03:14:20 2001
From: sobolak at myrealbox.com (Brian Sobolak)
Date: Tue, 23 Oct 2001 03:14:20 -0700
Subject: [buug] Re: [Buug-admin] Email List Rental
In-Reply-To: <E15vsH7-0000pa-00@neptune>
References: <0EAC4A3C6A28D51192C20090278733053505F3@PDC-WEB2WEB>
 <E15vsH7-0000pa-00@neptune>
Message-ID: <4442987042.20011023031420@myrealbox.com>

Hello C.,

Monday, October 22, 2001, 8:30:44 PM, you wrote:

CS> Please, please... those of you in other UNIX-oriented user groups... please 
CS> tell me that this sort of request is not common... and better yet, please 
CS> tell me that it is never honored...

CS> (Yes, I told the guy to get stiff.  Topic was his, BTW.  Rental?  No.)

I've never seen such a thing on any of the FreeBSD lists that I'm on,
and spam does leak through about once per day.  Perhaps it goes to the
postmaster and never makes it to the subscribers.

These guys are running Windoze too - check out the uptime stats...
<http://uptime.netcraft.com/up/graph/?host=www.acteva.com>

brian

--
Pull open the glass door, feel the rush of cool air, walk inside, get in
line, and look around you, look at the kids working in the kitchen, at the
customers in their seats, at the ads for the latest toys, study the
backlit color photographs above the counter, think about where the food
came from, about how and where it was made, about what it set in motion by
every single fast food purchase, the ripple effect near and far, think about
it.  Then place your order.  Or turn and walk out the door.  It's not too
late.  Even in this fast food nation, you can still have it your way.

        --Eric Schlosser, _Fast Food Nation_

Brian Sobolak                           San Francisco, California
sobolak at myrealbox.com                   http://www.planetshwoop.com





From fscked at pacbell.net  Tue Oct 23 04:39:44 2001
From: fscked at pacbell.net (richard childers)
Date: Tue, 23 Oct 2001 04:39:44 -0700
Subject: [buug] Re: [Buug-admin] Email List Rental
References: <0EAC4A3C6A28D51192C20090278733053505F3@PDC-WEB2WEB>
 <E15vsH7-0000pa-00@neptune>
Message-ID: <3BD55700.92C2C70F@pacbell.net>

Doesn't this depend upon the nature of the invitation?

-- richard

"C. Sullivan" wrote:

> On Monday 22 October 2001 16:31, you wrote:
> > dear BUUG administrator(s)
> >
> > I would like to promote a service to members of your user group and invite
> > them to an industry event. How can I do that?
> >
> > Thanks,
> >
> > Vincent Fallourd
> > Alliance Program Manager
> > (415) 374-8231
> > Acteva
> > 633 Battery Street, 2nd Floor
> > San Francisco, CA 94111
>
> Please, please... those of you in other UNIX-oriented user groups... please
> tell me that this sort of request is not common... and better yet, please
> tell me that it is never honored...
>
> (Yes, I told the guy to get stiff.  Topic was his, BTW.  Rental?  No.)
>
> -Fedl
>
> _______________________________________________
> Buug mailing list
> Buug at weak.org
> http://www.weak.org/mailman/listinfo/buug

--

Richard Childers
Sen[i]or UNIX administrator & chief bottle washer
phone: 415.664.6291
email: fscked at pacbell.net
radio: KG6HAC @ 145.150 mHz





From jammer at weak.org  Fri Oct 26 10:07:58 2001
From: jammer at weak.org (Jon McClintock)
Date: Fri, 26 Oct 2001 10:07:58 -0700
Subject: [buug] Free SS20 on Craigslist...
Message-ID: <20011026100758.A13281@weak.org>

Get it while it lasts:

  http://www.craigslist.org/sfo/sys/2021586.html

-Jon



From sobolak at myrealbox.com  Fri Oct 26 11:48:35 2001
From: sobolak at myrealbox.com (Brian Sobolak)
Date: Fri, 26 Oct 2001 11:48:35 -0700
Subject: [buug] Free SS20 on Craigslist...
In-Reply-To: <20011026100758.A13281@weak.org>
References: <20011026100758.A13281@weak.org>
Message-ID: <164333042880.20011026114835@myrealbox.com>

Hello Jon,

Friday, October 26, 2001, 10:07:58 AM, you wrote:

JM> Get it while it lasts:

JM>   http://www.craigslist.org/sfo/sys/2021586.html


Apparently there's a number of servers up for grabs at the Webvan
auction too.

brian





From jammer at weak.org  Fri Oct 26 11:46:41 2001
From: jammer at weak.org (Jon McClintock)
Date: Fri, 26 Oct 2001 11:46:41 -0700
Subject: [buug] Free SS20 on Craigslist...
In-Reply-To: <164333042880.20011026114835@myrealbox.com>; from sobolak@myrealbox.com on Fri, Oct 26, 2001 at 11:48:35AM -0700
References: <20011026100758.A13281@weak.org> <164333042880.20011026114835@myrealbox.com>
Message-ID: <20011026114641.A15056@weak.org>

On Fri, Oct 26, 2001 at 11:48:35AM -0700, Brian Sobolak wrote:
> Apparently there's a number of servers up for grabs at the Webvan
> auction too.

Yeah:

	http://www.dovebid.com/auctions/browsewebcast.asp?AuctionID=1058

-Jon



From sobolak at myrealbox.com  Sat Oct 27 21:24:07 2001
From: sobolak at myrealbox.com (Brian Sobolak)
Date: Sat, 27 Oct 2001 21:24:07 -0700
Subject: [buug] How is the Linux source code stored?
Message-ID: <104381990.20011027212407@myrealbox.com>

Hello

I just read in some article somewhere that the Linux kernel sources
aren't stored in CVS.  Not that this is the only source management
tool out there, but if the kernel hackers don't use CVS, what do they
use?  How does one gain commit access, etc?

brian





From bferrell at baywinds.org  Sun Oct 28 09:24:02 2001
From: bferrell at baywinds.org (Bruce Ferrell)
Date: Sun, 28 Oct 2001 09:24:02 -0800
Subject: [buug] How is the Linux source code stored?
References: <104381990.20011027212407@myrealbox.com>
Message-ID: <3BDC3F32.873196D0@baywinds.org>

Last time I looked at kernel.org CVS is and was used there.

Brian Sobolak wrote:
> 
> Hello
> 
> I just read in some article somewhere that the Linux kernel sources
> aren't stored in CVS.  Not that this is the only source management
> tool out there, but if the kernel hackers don't use CVS, what do they
> use?  How does one gain commit access, etc?
> 
> brian
> 
> _______________________________________________
> Buug mailing list
> Buug at weak.org
> http://www.weak.org/mailman/listinfo/buug

-- 
Bruce

One day at a time... One second if that's what it takes



From rick at linuxmafia.com  Sun Oct 28 11:53:14 2001
From: rick at linuxmafia.com (Rick Moen)
Date: Sun, 28 Oct 2001 11:53:14 -0800
Subject: [buug] How is the Linux source code stored?
In-Reply-To: <104381990.20011027212407@myrealbox.com>
References: <104381990.20011027212407@myrealbox.com>
Message-ID: <20011028115314.G4490@linuxmafia.com>

begin Brian Sobolak quotation:

> I just read in some article somewhere that the Linux kernel sources
> aren't stored in CVS.  Not that this is the only source management
> tool out there, but if the kernel hackers don't use CVS, what do they
> use?

They discuss matters against the Linux kernel mailing list, and diff
against kernel releases and interim patchlevels as announced on that
list.  Frequent releases prevent one's development targets from being
too far out of date, and the situation is slightly chaotic, but on the 
whole, it's tended to be productive chaos.

Various parties have suggested formal source-control regimes, which
Torvalds has thus far declined to adopt -- for specific reasons which
one can look up, but I haven't done so for this post.  There are
unofficial CVS archives, notably the one kernel hacker David Miller
maintains at vger.kernel.org .

Please see http://www.tux.org/lkml/ , for lots of details.

Patches get adopted through a somewhat Darwinian process on the mailing
list (whose name you will often see shortened to "lkml").  There is no 
formal structure of "committers".  The process is often criticised as
allegedly too unstructured -- notably by other groups that just chance
to use formal source control, and have clear boundaries between insiders
and outsiders according to access permissions.  Discussions of the
matter often turn out to be _not_ discussions, but rather attempts to 
foment flamewars between partisans.  I have no interest in the latter
and no investment in any position; I'm just trying to answer your
question as best I can.

-- 
Cheers,      "Transported to a surreal landscape, a young girl kills the first
Rick Moen     woman she meets, and then teams up with three complete strangers
rick at linuxmafia.com       to kill again."  -- Rick Polito's That TV Guy column,
              describing the movie _The Wizard of Oz_