Linux-Advocacy Digest #837, Volume #31 Tue, 30 Jan 01 02:13:04 EST
Contents:
Re: Linux headache ("Aaron R. Kulkis")
Re: Linux headache ("Aaron R. Kulkis")
Re: rh 6.2 and wu-ftp 2.6.0-14 ("Aaron R. Kulkis")
Re: Microsoft DEATH NECKLESS is COMPLETE!!! (Ed Allen)
Re: Linux Myths -- What I'd call Part II is here! (David Steinberg)
Re: M$ websites down again - Problem solved -> use Linux! ("Les Mikesell")
Re: M$ websites down again - Problem solved -> use Linux! ("Les Mikesell")
----------------------------------------------------------------------------
From: "Aaron R. Kulkis" <[EMAIL PROTECTED]>
Subject: Re: Linux headache
Date: Tue, 30 Jan 2001 01:41:59 -0500
Robert Morelli <[EMAIL PROTECTED]> wrote:
>
>
>
I guess the Utah's mathematic department must be scraping the bottom of
the barrel if one of their guys (even an undergrad) can't figure out
how to set up Linux off of a CD.
> In article <[EMAIL PROTECTED]>, "Andy Walker"
> <[EMAIL PROTECTED]> wrote:
>
> > I don't want to start a flame war over this but there is one thing I
> > want to get off my chest. Linux is an absolute nightmare to learn. Now
> <snip>
> > I'm just an average user who is sick of Micro$oft and I desparately want
> > Linux to succeed in the market place. However if the companies such as
> > RedHat, Mandrake etc don't address problems like these, a lot of the
> > momentum will be lost. All these companies seem to be spending all their
>
> Listen, we all want to see Linux succeed in the wider market place, but you
> have to be realistic. Linux is already gaining a lot of success in certain special
> server markets, but in a recent interview, Linus Torvalds predicted
> it would be 5 or 10 years before Linux challenges Microsoft on the
> desktop. I think it may have an impact earlier on corporate desktops
> that are administered by a dedicated staff. For the general public, I think
> you should just resign yourself to the fact that 5 years is an optimistic estimate
> for the general public, and 10 years is probably more realistic.
>
> Why?
> 1. Because your experience is absolutely typical.
> 2. Because Linux/Unix is based on very old fashioned principles. It has
> a certain elegance of its own; it was an elegant solution for the world it
> was born in, but that world is now a part of history. It's not a
> good fit for the modern world. A system like Windows was
> built with poor foresight, and horrendous design flaws. But in the
> ways that matter to end users, it is much more modern than Linux.
> 3. Because the Linux/Unix paradigm is designed for multiuser
> client/server environments, and this paradigm is totally useless for the
> typical home user. For instance, the root user under Linux can do
> things you should never do, while the non-root users can't do ordinary
> things you need to do. The root/non-root distinction in Linux is a
> nuissance which has no purpose for the home user.
> 4. Because it's the last 1% of the work that makes the difference between
> a smoothly working system and one that frustrates. But that last 1% is
> the most difficult and tedious to do and it rarely gets done in the sort
> of communal model Linux uses.
> 5. Because companies like Red Hat and Mandrake are largely just
> hackers who get a salary. What I mean by that is that these people
> typically have limited training and don't really understand the more
> subtle aspects of software development, and don't adhere to
> commercial quality standards and testing procedures. Until those
> companies start hiring a different breed of developer, the big Linux
> companies are going to just keep producing the same quality of
> software that you get from hackers on the internet.
> 6. Because user interface design is the most subtle and demanding aspect
> of programming. You can expect the Linux community to do simple
> things like write compilers (though gcc is at the moment in a pretty
> sorry state), but the open source model neither attracts the sort of
> talent necessary for user interface design, nor the level of direct interaction
> between users and programmers that is necessary to make it succeed.
>
> > time creating different installers for the O.S. but no one seems to be
> > doing much about making it more usable. May I be as bold as to suggest
> > someone creates a standard interface (or Wizard!) for decompressing then
> > building source code, which preferably involves no obscure commands or
> > syntax, installs the result and then puts it on the menu bar. Now I know
> > I'm not a programmer and I wouldn't know where to start with something
> > like this but I know every other O.S. does have simple to install
> > software and it's about time Linux did as well. Hope I haven't upset too
>
> You're right about this. Linux/Unix is a mess in this respect. What you
> may not realize is that Linux has already made big advances over the
> traditional Unix-think way of doing things. That way is to distribute
> software as tarballs of source code which a knowlegeable administrator
> unpacks, configures, compiles, and installs by spreading the binaries,
> libraries, and documentation into totally disorganized large system
> directories whose locations vary from vendor to vendor. The problem
> is that the tradition Unix-think way of doing things was about as
> stupid as if you'd set out deliberately to make it stupid.
>
> Linux has to overcome that stupidity before it can move forward to
> something intelligent. There is some work underway in that direction,
> like the Linux Standards Base, but don't hold your breath. It takes
> a lot of planning, cooperation, and intelligence to make something
> like this work, and those things are all in short supply in the Linux
> community. Things will probably gradually improve over time, but
> you probably won't have anything comparable to Windows for
> several years.
>
> > many people but it's in all our interests to see Linux succeed.......
>
> It's also in our interests to have world peace. But people keep starting
> wars.
>
> > Bye the way, has anyone who isn't a beardy geek (no insult intended)
> > actually successfully re-compiled their kernel and got exactly the
> > result they were looking for ???
>
> Sure, it's easy!
> 1. First, you download about 13 MBs of compressed source
> code.
> 2. Then you have to set about two hundred or so obscure parameters
> to create an appropriate make file. You'll probably make the false
> assumption the first time through that the defaults are appropriate.
> (Like the default: kernel hacking)
> 3. Then you build.
> 4. Then install the new kernel and rerun lilo.
> 5. Then reboot and get a kernel panic.
> 6. Then you go to another computer and post a desperate help message in the
> newsgroups.
> 7. A day or so later, you'll get your computer to boot again.
> 8. You go through the two hundred or so parameters more carefully,
> trying to figure out which ones were wrong.
> ...
> Trust me. After a few days, it finally will work.
>
> By the way, all kidding aside, asking the end user to do bizarre things
> like recompile the kernel, is one of many, many reasons why Linux
> is not appropriate for the end user. It reflects fundamental design
> flaws in Linux that need to be overcome before the system can
> compete with more technically advanced systems like Windows.
--
Aaron R. Kulkis
Unix Systems Engineer
DNRC Minister of all I survey
ICQ # 3056642
H: "Having found not one single carbon monoxide leak on the entire
premises, it is my belief, and Willard concurs, that the reason
you folks feel listless and disoriented is simply because
you are lazy, stupid people"
I: Loren Petrich's 2-week stubborn refusal to respond to the
challenge to describe even one philosophical difference
between himself and the communists demonstrates that, in fact,
Loren Petrich is a COMMUNIST ***hole
J: Other knee_jerk reactionaries: billh, david casey, redc1c4,
The retarded sisters: Raunchy (rauni) and Anencephielle (Enielle),
also known as old hags who've hit the wall....
A: The wise man is mocked by fools.
B: Jet Silverman plays the fool and spews out nonsense as a
method of sidetracking discussions which are headed in a
direction that she doesn't like.
C: Jet Silverman claims to have killfiled me.
D: Jet Silverman now follows me from newgroup to newsgroup
...despite (C) above.
E: Jet is not worthy of the time to compose a response until
her behavior improves.
F: Unit_4's "Kook hunt" reminds me of "Jimmy Baker's" harangues against
adultery while concurrently committing adultery with Tammy Hahn.
G: Knackos...you're a retard.
------------------------------
From: "Aaron R. Kulkis" <[EMAIL PROTECTED]>
Subject: Re: Linux headache
Date: Tue, 30 Jan 2001 01:44:35 -0500
Robert Morelli wrote:
>
> In article <[EMAIL PROTECTED]>, "Aaron R. Kulkis"
> <[EMAIL PROTECTED]> wrote:
>
> > Robert Morelli wrote:
> >>
> >> In article <[EMAIL PROTECTED]>, "Andy Walker"
> >> <[EMAIL PROTECTED]> wrote:
> >>
> >> > I don't want to start a flame war over this but there is one thing I
> >> > want to get off my chest. Linux is an absolute nightmare to learn.
> >> > Now
> >> <snip>
> >> > I'm just an average user who is sick of Micro$oft and I desparately
> >> > want Linux to succeed in the market place. However if the companies
> >> > such as RedHat, Mandrake etc don't address problems like these, a lot
> >> > of the momentum will be lost. All these companies seem to be spending
> >> > all their
> >>
> >> Listen, we all want to see Linux succeed in the wider market place,
> >> but you have to be realistic. Linux is already gaining a lot of
> >> success in certain special server markets, but in a recent interview,
> >> Linus Torvalds predicted it would be 5 or 10 years before Linux
> >> challenges Microsoft on the desktop. I think it may have an impact
> >> earlier on corporate desktops that are administered by a dedicated
> >> staff. For the general public, I think you should just resign
> >> yourself to the fact that 5 years is an optimistic estimate for the
> >> general public, and 10 years is probably more realistic.
> >>
> >> Why?
> >> 1. Because your experience is absolutely typical.
> >> 2. Because Linux/Unix is based on very old fashioned principles. It
> >> has a certain elegance of its own; it was an elegant solution for the
> >> world it was born in, but that world is now a part of history. It's
> >> not a good fit for the modern world. A system like Windows was built
> >> with poor foresight, and horrendous design flaws. But in the ways
> >> that matter to end users, it is much more modern than Linux.
> >> 3. Because the Linux/Unix paradigm is designed for multiuser
> >> client/server environments, and this paradigm is totally useless for
> >> the typical home user. For instance, the root user under Linux can do
> >> things you should never do, while the non-root users can't do ordinary
> >> things you need to do. The root/non-root distinction in Linux is a
> >> nuissance which has no purpose for the home user.
> >> 4. Because it's the last 1% of the work that makes the difference
> >> between a smoothly working system and one that frustrates. But that
> >> last 1% is the most difficult and tedious to do and it rarely gets done
> >> in the sort of communal model Linux uses.
> >> 5. Because companies like Red Hat and Mandrake are largely just
> >> hackers who get a salary. What I mean by that is that these people
> >> typically have limited training and don't really understand the more
> >> subtle aspects of software development, and don't adhere to commercial
> >> quality standards and testing procedures. Until those companies start
> >> hiring a different breed of developer, the big Linux companies are
> >> going to just keep producing the same quality of software that you get
> >> from hackers on the internet.
> >> 6. Because user interface design is the most subtle and demanding
> >> aspect of programming. You can expect the Linux community to do simple
> >> things like write compilers (though gcc is at the moment in a pretty
> >> sorry state), but the open source model neither attracts the sort of
> >> talent necessary for user interface design, nor the level of direct
> >> interaction between users and programmers that is necessary to make it
> >> succeed.
> >>
> >> > time creating different installers for the O.S. but no one seems to
> >> > be doing much about making it more usable. May I be as bold as to
> >> > suggest someone creates a standard interface (or Wizard!) for
> >> > decompressing then building source code, which preferably involves no
> >> > obscure commands or syntax, installs the result and then puts it on
> >> > the menu bar. Now I know I'm not a programmer and I wouldn't know
> >> > where to start with something like this but I know every other O.S.
> >> > does have simple to install software and it's about time Linux did as
> >> > well. Hope I haven't upset too
> >>
> >> You're right about this. Linux/Unix is a mess in this respect. What
> >> you may not realize is that Linux has already made big advances over
> >> the traditional Unix-think way of doing things. That way is to
> >> distribute software as tarballs of source code which a knowlegeable
> >> administrator unpacks, configures, compiles, and installs by
> >> spreading the binaries, libraries, and documentation into totally
> >> disorganized large system directories whose locations vary from vendor
> >> to vendor. The problem is that the tradition Unix-think way of doing
> >> things was about as stupid as if you'd set out deliberately to make it
> >> stupid.
> >>
> >> Linux has to overcome that stupidity before it can move forward to
> >> something intelligent. There is some work underway in that direction,
> >> like the Linux Standards Base, but don't hold your breath. It takes a
> >> lot of planning, cooperation, and intelligence to make something like
> >> this work, and those things are all in short supply in the Linux
> >> community. Things will probably gradually improve over time, but you
> >> probably won't have anything comparable to Windows for several years.
> >>
> >> > many people but it's in all our interests to see Linux succeed.......
> >>
> >> It's also in our interests to have world peace. But people keep
> >> starting wars.
> >>
> >> > Bye the way, has anyone who isn't a beardy geek (no insult intended)
> >> > actually successfully re-compiled their kernel and got exactly the
> >> > result they were looking for ???
> >>
> >> Sure, it's easy!
> >> 1. First, you download about 13 MBs of compressed source
> >> code.
> >> 2. Then you have to set about two hundred or so obscure parameters
> >> to create an appropriate make file. You'll probably make the false
> >> assumption the first time through that the defaults are appropriate.
> >> (Like the default: kernel hacking)
> >> 3. Then you build.
> >> 4. Then install the new kernel and rerun lilo.
> >> 5. Then reboot and get a kernel panic.
> >> 6. Then you go to another computer and post a desperate help message
> >> in
> >> the newsgroups.
> >> 7. A day or so later, you'll get your computer to boot again.
> >> 8. You go through the two hundred or so parameters more carefully,
> >> trying to figure out which ones were wrong.
> >> ...
> >> Trust me. After a few days, it finally will work.
> >>
> >> By the way, all kidding aside, asking the end user to do bizarre
> >> things like recompile the kernel, is one of many, many reasons why
> >> Linux is not appropriate for the end user. It reflects fundamental
> >> design flaws in Linux that need to be overcome before the system can
> >> compete with more technically advanced systems like Windows.
> >
> > All of this must be why UNIX is the STANDARD desktop for every
> > automotive designer in the world...all the way from the manufacturers
> > (GM, FORD, BMW, VW, TOYOTA, ROLLS ROYCE, etc.) all the way down to the
> > most insignificant tier-3 parts supplier.
>
> Actually, UNIX has been the standard in university departments,
> where I've been for about 20 years. Like I said in my post, I think
> It's entirely reasonable to run UNIX on the desktop in an environment
> where the maintainance is done by a dedicated staff.
>
> Even so, a lot of people have always used Macs in this environment,
> because the usability and software availability of UNIX has always
> been so poor. During the 90's, the PC platform also became
> increasingly popular, as Windows became gradually less and less
> brain dead. From what I hear, Windows 2000 may actually have
> a functioning conscious brain, but I don't currently use any version
> of Windows so I can't say.
>
> The emergence of Linux has actually brought about a resurgence
> of UNIX-like systems for desktop use in academia. However,
> academia is a special environment. Sentiment is almost univerally
> strongly anti-Microsoft here, people tend to be familiar with
> UNIX here, and they are generally more willing to deal with crude
> and experimental systems like Linux.
>
> I don't think special environments like the automotive industry and
> academia really reflect what the general public is willing to deal with.
> Linux and UNIX have some bright spots, and can function reliably
> under controlled circumstances, but they are much too barbaric
> for the typical home user.
The ONLY reason MS-DOS ever found it's way into American homes was because
managers wanted to use the same thing at home as what they had at work.
When the corporate desktop goes to Linux, the American household
will soon follow.
Why? Because those will be the homes of MANAGEMENT...i.e. THE $$$$$ households.
--
Aaron R. Kulkis
Unix Systems Engineer
DNRC Minister of all I survey
ICQ # 3056642
H: "Having found not one single carbon monoxide leak on the entire
premises, it is my belief, and Willard concurs, that the reason
you folks feel listless and disoriented is simply because
you are lazy, stupid people"
I: Loren Petrich's 2-week stubborn refusal to respond to the
challenge to describe even one philosophical difference
between himself and the communists demonstrates that, in fact,
Loren Petrich is a COMMUNIST ***hole
J: Other knee_jerk reactionaries: billh, david casey, redc1c4,
The retarded sisters: Raunchy (rauni) and Anencephielle (Enielle),
also known as old hags who've hit the wall....
A: The wise man is mocked by fools.
B: Jet Silverman plays the fool and spews out nonsense as a
method of sidetracking discussions which are headed in a
direction that she doesn't like.
C: Jet Silverman claims to have killfiled me.
D: Jet Silverman now follows me from newgroup to newsgroup
...despite (C) above.
E: Jet is not worthy of the time to compose a response until
her behavior improves.
F: Unit_4's "Kook hunt" reminds me of "Jimmy Baker's" harangues against
adultery while concurrently committing adultery with Tammy Hahn.
G: Knackos...you're a retard.
------------------------------
From: "Aaron R. Kulkis" <[EMAIL PROTECTED]>
Subject: Re: rh 6.2 and wu-ftp 2.6.0-14
Date: Tue, 30 Jan 2001 01:47:21 -0500
Tom Wilson wrote:
>
> "Chad" <[EMAIL PROTECTED]> wrote in message
> news:Coid6.39598$[EMAIL PROTECTED]...
> > I have rh 6.2 and recently installed package wu-ftpd-2.6.0-14.6x. My
> > security team says, "Version 2.6.0 still contains security exposures".
> They
> > are suggesting to install wu 2.6.1, but I would rather just stick with
> the
> > rh rpms. Can someone shed some light on this?
>
> The only light that needs shed is that your security guys are right. Listen
> to them.
>
> Keep that package installed and you can expect the box to be rooted in no
> time. Give me the URL to it and I'll be happy to show you. Its' a commonly
> known weakpoint.
>
> PS: Red Hat doesn't maintain that package. Its' an add on included with
> your distro.
>
> PSS: Listen to your security team!
Chad NEVER LISTENS!
(That's why the ignorant twit keeps singing the praises of Microsoft's
rediscovery of 1960's methodologies).
>
> --
> Tom Wilson
> Sunbelt Software Solutions
--
Aaron R. Kulkis
Unix Systems Engineer
DNRC Minister of all I survey
ICQ # 3056642
H: "Having found not one single carbon monoxide leak on the entire
premises, it is my belief, and Willard concurs, that the reason
you folks feel listless and disoriented is simply because
you are lazy, stupid people"
I: Loren Petrich's 2-week stubborn refusal to respond to the
challenge to describe even one philosophical difference
between himself and the communists demonstrates that, in fact,
Loren Petrich is a COMMUNIST ***hole
J: Other knee_jerk reactionaries: billh, david casey, redc1c4,
The retarded sisters: Raunchy (rauni) and Anencephielle (Enielle),
also known as old hags who've hit the wall....
A: The wise man is mocked by fools.
B: Jet Silverman plays the fool and spews out nonsense as a
method of sidetracking discussions which are headed in a
direction that she doesn't like.
C: Jet Silverman claims to have killfiled me.
D: Jet Silverman now follows me from newgroup to newsgroup
...despite (C) above.
E: Jet is not worthy of the time to compose a response until
her behavior improves.
F: Unit_4's "Kook hunt" reminds me of "Jimmy Baker's" harangues against
adultery while concurrently committing adultery with Tammy Hahn.
G: Knackos...you're a retard.
------------------------------
From: [EMAIL PROTECTED] (Ed Allen)
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.ms-windows.nt.advocacy
Subject: Re: Microsoft DEATH NECKLESS is COMPLETE!!!
Reply-To: [EMAIL PROTECTED]
Date: Tue, 30 Jan 2001 07:00:09 GMT
In article <[EMAIL PROTECTED]>,
Peter Hayes <[EMAIL PROTECTED]> wrote:
>
>Will the GPL stop Microsoft from packaging a Linux distro, adding their own
>bits and flogging it to the masses under the banner of Microsoft Linux?
>
No but then Linux from other vendors would be just as acceptable as
the Microsoft one.
Any benchmark claims or TCO claims will be compared directly to
RedHat, SuSE, and TurboLinux. Even the Microsoft advertising
dollars would not be able to keep all the comparisons hidden.
How do they recoup the money they sunk into NT5 (Win2000) when their
own version of software designed and written elsewhere demonstrates
several times the reliability on the same hardware ?
No MSLinux would allow competition wherever they marketed it. They
do not want competition.
Bill Gates has said that he can only see making money if he has what
he calls a "singularity." Single anythings do not allow for
comparison.
>With the Microsoft "seal of approval" the OEMs will fall over themselves to
>offer this "New Microsoft OS" preload to the public who in turn will rush
>out lemming like to buy it, thereby reviving the flagging PC market.
>
The contracts Microsoft has with "Windows only" hardware might be
harder to renew.
Hardware makers would see themselves limited to Windows while the
system vendor would have a larger market.
>And the only stuff M$ have to GPL is what they've lifted from other
>distros, they don't have to GPL anything they wrote themselves (like SuSE's
>Yast, or Id's games).
>
Nothing wrong with this but they must begin with a clean slate and
then develop something better than SuSE without the years of polish
that the other vendors have on their installs.
Remember how many years it took ISVs to develop apps which were
better than the bundled ones written by Microsoft programmers who
had two years experience writing Windows software before they did.
Gates is a strong believer in the concept of being first with
software because followers need to have much better quality and
features to dislodge the earlier system.
Microsoft would be conceding that first adopter spot to others.
>So you have Microsoft Linux for the SOHO user, Whistler for the Enterprise,
>and Microsoft Office for Linux plus Microsoft Wine as the bridge between
>the two.
>
What price would you expect MSLinux to sell for ?
If, as you seem to imply, it would be around $189 what happens to ME ?
Microsoft would still have the problem of having a Microsoft blessed
solution which costs one fourth as much but has all the features of
Enterprise and better reliability.
Competition is the last thing a monopoly wants.
--
FYI. When you do type "make" on the Windows NT source tree, it takes almost
38 hours for it to complete on a 4-way 400 Mhz PII System, as opposed to
about 5 minutes on Linux. Linux is not Doomed!!!!!! -- Jeff Merkey
http://boudicca.tux.org/hypermail/linux-kernel/1999/1999week26/0787.html
------------------------------
From: [EMAIL PROTECTED] (David Steinberg)
Subject: Re: Linux Myths -- What I'd call Part II is here!
Date: 30 Jan 2001 07:00:26 GMT
Salvador Peralta ([EMAIL PROTECTED]) wrote:
: David Steinberg wrote:
: > --
: > David Steinberg -o)
: > Computer Engineering Undergrad, UBC / \
: > [EMAIL PROTECTED] _\_v
: ^^^^^
: Nice ascii
: is it GPL'd?
I don't know. A few people were using it on a local LUG mailing list. I
really liked it, and wanted to put it my own .sig, so of course I asked
who created it and if it could be used freely. Nobody knew. I suppose
the responsible thing would have been to track down the answer, but I just
started using it. :)
--
David Steinberg -o)
Computer Engineering Undergrad, UBC / \
[EMAIL PROTECTED] _\_v
------------------------------
From: "Les Mikesell" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.nt.advocacy
Subject: Re: M$ websites down again - Problem solved -> use Linux!
Date: Tue, 30 Jan 2001 07:01:36 GMT
"CR Lyttle" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> > >
> > > Tech support did *NOT* start getting calls immediately. DNS caches
take a
> > > great deal of time to expire, usually 24 hours or more.
> >
> > That is only for someone who had loaded the record just before the
> > link was broken. A site that hadn't accessed it for a while would
> > fail immediately. If people weren't calling, it was only because
> > they knew from experience that it wouldn't help... Note that the
> > problem was well known on slashdot and similar sites in the morning,
> > but it wasn't fixed until evening. And all it takes is a traceroute to
> > diagnose a problem like that, and a few minutes to correct the router
> > config.
> I don't think anyone at MS noticed until /. posted the story. I'll admit
> it takes up to 24 hours for the whole world to not be able to find MS
> and during that time you could still ping many of the sites with the
> dotted quad. That was what I was told to do when I tried to access a
> site served by MS. I called in a report, was told to try to ping. I
> pinged, got back a response (330 ms delay, 60% dropped) and was told,
> "See we're up, its your Domain Name resoulution."
Of course it was DNS, but it wasn't yours, and it shouldn't have been
up to you to prove it. Slashdot had the story in the morning. How
long did it take them to figure out how they had misspelled traceroute
and try it?
Les Mikesell
[EMAIL PROTECTED]
------------------------------
From: "Les Mikesell" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.nt.advocacy
Subject: Re: M$ websites down again - Problem solved -> use Linux!
Date: Tue, 30 Jan 2001 07:06:32 GMT
"Erik Funkenbusch" <[EMAIL PROTECTED]> wrote in message
news:Czsd6.335$[EMAIL PROTECTED]...
> "Charles Lyttle" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > > You aren't listening. Since the failure occured a great deal of time
> after
> > > the change was made, it was difficult to connect the failure with that
> > > specific change. It took them a great deal of time to track down
> exactly
> > > where the problem was. Once they knew where it was, solving it was
> quite
> > > easy.
>
> > So how often do they have to make changes to their software? How many
> > changes had to be backed out to get the working configuration back? What
> > exactly was the change that caused the problem? If it wasn't the fault
> > of MS software, why isn't the change published with a "DON'T DO THIS"
> > warning?
>
> One more time. It was a router configuration problem with the router
> (probably a CISCO). MS didn't notice the problem internally because the
> router was only failing to route incoming packets, and local packets were
> routing just fine.
>
> A site as big and busy as MS is has hundreds or thousands of changes done
to
> it daily.
That doesn't leave much excuse for putting both nameservers on the
same subnet, does it?
> Figuring out a problem requires testing an awful lot of things
> before even narrowing it down to a specific piece of hardware, and a
> specific configuration. You don't know if it's a hardware failure, a
> configuraiton failure, a NAP failure, etc..
Traceroute tells you where the packets stop. Fix that box and you
are usually done.
Les Mikesell
[EMAIL PROTECTED]
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.advocacy.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Advocacy Digest
******************************
