Linux-Misc Digest #367, Volume #19 Mon, 8 Mar 99 09:13:15 EST
Contents:
Re: Moving /home to /usr/home ("David Z. Maze")
Linux Safeguards for the Consumer? ("Benjamin Sher")
Re: windows 95B doesn't see FAT32 partition (Fred Heitkamp)
Re: best offline newsreader? ("Richard Latimer")
Re: Kernel NFS Problem; device busy (Rainer Krienke)
Re: windows 95B doesn't see FAT32 partition (Fred Heitkamp)
Re: tar question (John Thompson)
Re: Best value in CPU for linux (Markus Wandel)
Re: OS with a seamless object model (Bjorn Borud)
Re: Linux Safeguards for the Consumer? (**Nick Brown)
Re: No-Win Modem Situation (Andrew Comech)
xosview and kernel 2.2.2 (Jean-Yves TOUMIT)
who on RH5.1 (new machine) (Rick Lim)
Re: Pentium III Boycott and survey info (bp jendrissek)
Re: crypt() linking error (J.H.M. Dassen (Ray))
Re: Linux Safeguards for the Consumer? (Rod Smith)
Problem with shadow - redhat 5.2 - kernel 2.2 (Jean-Yves TOUMIT)
----------------------------------------------------------------------------
From: "David Z. Maze" <[EMAIL PROTECTED]>
Subject: Re: Moving /home to /usr/home
Date: 08 Mar 1999 07:59:01 -0500
gbh <[EMAIL PROTECTED]> writes:
gbh> I'd like to move my /home directory to /usr/home and would like
gbh> to know the correct way to do it and what problems this may
gbh> create.
(Why? The general idea with /usr is that it's data that doesn't
change unless you reinstall or upgrade your distribution, with the
possible exception of /usr/local, and /usr can (should?) be mounted
read-only. Plus AFAIK there's no legacy software out there that
expects to find user directories under /usr/home. If it's "I have a
huge /usr partition but no other space on my disk," and
gbh> I've just installed RH 5.2 and have not setup any user accounts
gbh> yet.
then you're probably better off reinstalling and either creating a
sizeable /home partition or not splitting off /usr.)
gbh> I created a small partition for the / directory and a large
gbh> partition for the /usr directory. Should I first delete /home,
gbh> then create /usr/home, then create a symbolic link for example
gbh> 'ln -s /usr/home /home'?
If you were going to do this, that's pretty much the right way, with a
check that /usr/home has the same permissions as the original /home.
gbh> What other directories should be moved from / to /usr and linked
gbh> back to /?
None of them.
gbh> And how large should the partition be to hold the / directory? My
gbh> / directory occupies about 30 MB. In future installations, if I
gbh> set the partition to 50 MB would it ever grow to more than this?
This really depends on how you've partitioned your hard drive and what
you install. If your setup is
/ 50 MB
/usr (a lot)
then you'll be hurting for space in the root partition; my Debian
package database, among other things, lives in /var (and I believe RPM
does the same thing), and you'll need more room in /home. If your
setup is more like mine:
/ 32 MB
/usr 1000 MB
/var 300 MB
/home 400 MB
/usr/local 500 MB
(with /tmp being a symlink on to /var, and on three separate disks,
but no other weirdnesses) you'll be fine for space on the root
partition.
--
David Maze [EMAIL PROTECTED] http://donut.mit.edu/dmaze/
"Hey, Doug, do you mind if I push the Emergency Booth Self-Destruct Button?"
"Oh, sure, Dave, whatever...you _do_ know what that does, right?"
------------------------------
From: "Benjamin Sher" <[EMAIL PROTECTED]>
Subject: Linux Safeguards for the Consumer?
Date: Mon, 08 Mar 1999 12:46:48 GMT
Dear friends:
As a consumer with some experience with Dos, Win3.1 and Win95, I am in the
process of deciding whether I should switch over to Linux. I am considering
buying Linux on a Disk (LOAD) made by Cosmos Engineering
(www.cosmoseng.com). I have been reading deeply in Red Hat Linux Unleashed
in order to help me understand Linux and make a wise decision.
Specs: NEC Pentium 166, MMX, 64 meg Ram. Win95 + Linux (hopefully) using
LILO.
QUESTIONS:
1 -- ROOT MODE SAFEGUARDS:
Are there any safeguards (either command-line or by way of the GUI) against
accidentally hitting the wrong key combination while in "root" mode? Does
hitting such a combination of keys really result in destruction of your
system (as opposed to "merely" requiring a reinstallation of Linux?
2 -- SHUTDOWN SAFEGUARDS:
I am planning to buy a battery to protect me from power failures, etc.
Makes a lot of sense. But what about human error? Inadvertently hitting the
power button to shut down the system. I understand that fsck or File System
Checker is the LInux equivalent of Scan Disk. Is it a sufficient safeguard
against human error?
I have read that Linux, unlike Windows, runs in "unprotected" mode. What
exactly does that mean, from a practical point of view? Does that mean that
the safeguards above do not apply or cannot really be applied to Linux?
My decision will depend on the answers I get from you experts.
Another way to put is: Is Linux ready for the consumer?
Thank you very much.
Benjamin Sher
[EMAIL PROTECTED]
------------------------------
Date: Mon, 08 Mar 1999 07:59:42 -0500
From: Fred Heitkamp <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.misc
Subject: Re: windows 95B doesn't see FAT32 partition
MINIMAN wrote:
> You are probably using FAT16 in your windows 95 B
>
> If you aren't, try formating your partition through windows as FAT32,
> I am sure you will see it then
The "plot" is thickening. I just noticed that the drive made with mkdosfs
is "seen" by Windows, it's just shown as the F: drive not the E: drive as
intended. I can highlight the E: drive icon, and presumably format it.
One would assume the F: drive would then go away. Wierd.
Fred
------------------------------
From: "Richard Latimer" <[EMAIL PROTECTED]>
Subject: Re: best offline newsreader?
Date: Mon, 8 Mar 1999 05:12:29 -0800
Thanks to everyone who has jumped in here and expressed themselves
concerning my diatribe.
Before I write anything else, a small disclaimer is in order. I did not mean
to imply that we should introduce html usage into newsgroups. I only meant
to point out that Outlook Express, a combined mail and news client, has the
ability to handle about anything you can send it, or wish to create with it.
If you had a new baby, you could cobble up some dynamite announcements
in no time, with cutsey pink and blue borders, pics of the kid, and some
gurgling sounds to go with it. That's a more practical html usage. The
functionality is there when you need it. And it's fun (excepting the guys
in the hair shirts).
=====
The posts in this thread are interesting. Notice how many users are not
running all linux systems, or have additional machines running other
operating systems. Obviously, Linux cannot meet all of their computing
needs, yet. Excluding server installations, there are probably few linux
users who do not have a second operating system loaded, or a second
computer running another operating system.
The question is why?
One answer is economic viability. Bell Labs never exploited unix as
a commercial consumer product. It was always too expensive for the
average user. It found a home on servers and high powered workstations.
A retail consumer base just never developed. Retail hardward and software
support similar to Microsoft's never developed because the user base
wasn't there.
When Torvald introduced Linux, all of that changed. It is now possible for
unix (Linux) to challenge Windows on the desktop. The necessary work
for success, however, has just not been done. The foreseeable future is a
world of Linux servers serving Windows clients.
Why?
Getting information from hardware manufacturers is difficult. The suite
of support software needed by the kernel requires a lot of effort. But
there is an additional reason. It is the unix ethos that has survived in the
Linux community. Reading the Linux newsgroups I am reminded of that
cartoon about the Plugger. You know the one about the big bear in the
flannel shirt, wearing a baseball cap. Multimedia to him would be a pinup
of Betty Grable and a transistor radio he bought in the fifties.
On my RedHat installation I was given elm, exmh, fetchmail, mailx,
metamail, mutt, nmh, pine, slrn, tin, trn, mailcap, procmail, sendmail,
ppp, dip, netscape-communicator, xrn, and xmailbox. If I list the support-
ing net packages, the list more than doubles. All but a few of these
packages can be discarded, but to a newbit, which? Which rpm removal
will break the setup? RTFM, find the HOWTo. Uh-huh. All of this just
to get some pop 3 mail and some nntp messages?
In Windows I was given Outlook Express. In Windows I was asked for email
address, smtp, pop3, and nntp, my user id, my password, and the phone
number. Done. It didn't ask for DNS. It didn't ask me about ppp, chap, or
pap. It didn't ask me about the modem, it seemed to know where it was and
how to handle it. I didn't need to learn a new user interface, I already
knew
how to run a Windows app. It wasn't necessary to develop a keyboard
template, like the ones that were available for WordPerfect, so I could
remember the relevant keystrokes.
What seems reasonable in the unix community looks positively Luddite
from the outside. By accepting these tools as they are, by adapting your-
self to a rather Byzantine manner of doing things, you are undermining
the development that would make Linux a winner on the desktop. Urging
others to accept this culture as reasonable is a mistake.
Linux needs the desktop user base. More users mean better hardware
and software support, better quality, fuller featured apps, and more money
for taming the beast by the distributors.
Experienced linux users are needed for the desktop development. They are
in a better position to help than newbits who have just fleed Microsoft.
Their
bug reports will contain better information for developers. Their
development
suggestions will be in greater sync with the Linux system.
If experienced users are content with the arcana, then they are not
interested
in the development that is needed and not participating in the necessary
change.
There is only a limited opportunity for Linux to succeed on the desktop.
Look
at what has happened to Microsoft. They frustrated users. The frustration
lead to anger. The anger lead to contempt, and the contempt lead to hatred.
Linux is very frustrating for the uninitiated. The operating system doesn't
crash, the apps one wants to use core bomb. It provides flexible
configuration,
but is very difficult to configure. It does provide a home for irate Windows
users, but only after they have checked their niftiest hardward and software
at the door. It won't take much for Linux to develop a bad rep among
desktop users.
You can see common reactions to Linux here. The newsgroups always
contain some frustrated newbit ranting in expletives about Linux. Pay
attention
to what they are saying. They are Linux's future. To successfully challenge
Microsoft's hegemony Linux needs to adapt to them. They are not going
to adapt to Linux.
Generous people wrote a lot of unix/linux software to solve specific
problems. It was adequate for their purposes. It is no longer adequate for
the challenges Linux confronts now. More elegant, simpler solutions are
needed.
richard
------------------------------
From: [EMAIL PROTECTED] (Rainer Krienke)
Subject: Re: Kernel NFS Problem; device busy
Crossposted-To: comp.os.linux.development.system
Date: 8 Mar 1999 14:20:00 +0100
[Posted and mailed]
In article <7bl594$7v2$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Tim Dawson) writes:
> If you exported it (required for remote systems to mount the partition)
> then the mount point truly IS busy (by the NFS daemons and/or mountd) and
> hence cannot be unmounted. Try unexporting it first.
>
This is bad habit. If you only export it, there may be no one that
really mounts ans useses the exported directory. So it should only be
busy if someone really has a NFS mount sitting on it.
Rainer
Krienke
--
=====================================================================
Rainer Krienke [EMAIL PROTECTED]
Universitaet Koblenz, http://www.uni-koblenz.de/~krienke
Rechenzentrum, Voice: +49 261 287 - 1312
Rheinau 1, 56075 Koblenz, Germany Fax: +49 261 287 - 1355
=====================================================================
------------------------------
Date: Mon, 08 Mar 1999 07:52:20 -0500
From: Fred Heitkamp <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.misc
Subject: Re: windows 95B doesn't see FAT32 partition
MINIMAN wrote:
> You are probably using FAT16 in your windows 95 B
>
> If you aren't, try formating your partition through windows as FAT32,
> I am sure you will see it then
The "plot" is thickening. I just noticed that the drive made with mkdosfs
is "seen" by Windows, it's just shown as the F: drive not the E: drive as
intended. I can highlight the E: drive icon, and presumably format it.
One would assume the F: drive would then go away. Wierd.
Fred
------------------------------
From: John Thompson <[EMAIL PROTECTED]>
Subject: Re: tar question
Date: Sun, 07 Mar 1999 21:40:47 -0600
Ron wrote:
>
> I have a harddrive that is almost full. It contains
> alot of small files (thousands) that I need to backup
> to tape, is there a tar command that will allow me to
> get all these files into a single archive and onto the
> tape without having the available space to creat it
> on the disk first?
AFAIK, tar does not create the archive on disk and then move
it to the tape; it writes its output directly to the tape
device (or pipes it through gzip on the way to the tape).
At least it sure seems to work that way here...
--
-John ([EMAIL PROTECTED])
------------------------------
From: [EMAIL PROTECTED] (Markus Wandel)
Subject: Re: Best value in CPU for linux
Date: 8 Mar 1999 13:14:44 GMT
I didn't own a PC at all until two months ago, and I wanted one to run Linux.
Strategy to use? Stick with the "standard hardware." And what is standard
hardware? Any current chipset by Intel that has been out for more than 6
months or so. It is inconceivable that a PC operating system can have
incompatibilities with such a chip set for very long.
I simply heard enough horror stories about AGP, UDMA etc. not _quite_ working
with clone chipsets unless proprietary "drivers" were installed, and then there
is USB which everyone has but hardly anyone uses yet. I am much more
confident that if I suddenly need USB it will work with a 2-year-old Intel
chip set than with an equivalent clone chip set.
So we have the chip set: Intel 440BX. Everything else on the motherboard is
glue logic, so take your pick. I bought an ABIT BH6 on the recommendation
of the guy I bought from. Oh yes, be very careful with motherboards that have
built-in sound/video because you'll be stuck with it.
Now you have a slot 1 system, what to plug into your slot 1? The cheapest
thing was a Celeron 300A. What I didn't realize is just how kick-ass fast
this CPU is. Yes, it's 1.5 times as fast yet _if_ you can overclock it to
450MHz, but it's still faster -- a lot -- than the Pentium 233MMX systems
that your friends have. Of course in a year there will be faster systems yet.
I can't believe how much compute power I got for my buck, all while supporting
Intel's virtual monopoly. I believe I benefitted from some cut-throat pricing
policy designed to kill AMD -- who wants a K6 when you can get a faster Intel
processor for less money? I don't feel entirely good about this but that's
life.
The result? Win95 and RH5.2 Linux came right up with _zero_ problems since.
No weird drivers had to be installed or anything. I'm happy.
My opinion. Like a**holes, everyone has one. Don't flame me if yours
is different.
Markus
------------------------------
From: Bjorn Borud <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.linux.advocacy
Subject: Re: OS with a seamless object model
Date: 08 Mar 1999 14:27:15 +0100
[Francois-Rene Rideau <[EMAIL PROTECTED]>]
|
| Files suck anyway.
that may be true, but files are an extremely simple concept and most
people have the mental aparatus to deal with them.
-Bj�rn
--
Bj�rn Borud <[EMAIL PROTECTED]>, <URL:http://borud.priv.no/>
------------------------------
From: **Nick Brown <[EMAIL PROTECTED]>
Subject: Re: Linux Safeguards for the Consumer?
Date: Mon, 08 Mar 1999 14:27:36 +0100
Reply-To: [EMAIL PROTECTED]
It's no more or less ready for the "consumer" (whoever that is) than
DOS/Windows, which includes powerful tools (FDISK, FORMAT, REGEDIT,
recursive deletes from the top of the tree in Explorer, etc etc) which
let you screw up your installation.
And of course there's the inevitable "one bad block out of 18 million on
the disk". You have perhaps one chance in ten of a bad block taking out
something relatively important on your system over a three-year period,
I would guess. Nothing to do with the software you use.
"Destruction of your system" ? No. I mean, deleting every file would
count as destruction for me, but there are no programs called
"run-disk-at-99999-rpm-until-it-breaks" or
"overclockCPUuntil-smoke-detector-sounds".
If you think you might accidentally wipe out your system, my advice
would be to make sure that the first thing you learn is how to restore
complete backups. You will need this knowledge one day (on any system).
>From a marketing point of view, Linux is honest enough to tell you that
there is no warranty. M$ like you to think that they stand behind their
products, but in practice the chances of you actually getting a bug
fixed are vanishingly small.
Benjamin Sher wrote:
>
> Dear friends:
>
> As a consumer with some experience with Dos, Win3.1 and Win95, I am in the
> process of deciding whether I should switch over to Linux. I am considering
--
===============================================================
|\ | o _ |/ Life's like a jigsaw
| \| | |_ |\ You get the straight bits
But there's something missing in the middle
Nick Brown, Strasbourg, France (Nick(dot)Brown(at)coe(dot)fr)
===============================================================
------------------------------
From: Andrew Comech <[EMAIL PROTECTED]>
Subject: Re: No-Win Modem Situation
Crossposted-To: alt.os.linux,comp.os.linux.setup
Date: 8 Mar 1999 08:35:44 -0500
There is a (short so far) list of places which may have
hardware-based internal V.90 modems:
http://www.math.sunysb.edu/~comech/tools/CheapBox.html#modem
HTH,
Andrew
------------------------------
From: Jean-Yves TOUMIT <[EMAIL PROTECTED]>
Subject: xosview and kernel 2.2.2
Date: Mon, 08 Mar 1999 13:39:44 +0000
Hi all.
I just upgraded to kernel 2.2.2 (I had the 2.2.1 before) and suddenly,
xosview doesn't work anymore. To recompile the new kernel, I just copied
the .config from the old kernel directory to the new one.
I have verified that the /proc filesystem is included in it, it seems
OK.
What happens is that xosview runs and keeps on running using 80% CPU but
without displaying any window...
Any ideas? (BTW, any other good programs that would be equivalent to
xosview?)
--
Jean-Yves TOUMIT
mailto:[EMAIL PROTECTED]
http://rfv-pc28.insa-lyon.fr
------------------------------
From: Rick Lim <[EMAIL PROTECTED]>
Subject: who on RH5.1 (new machine)
Date: 8 Mar 1999 13:43:35 GMT
I have just installed RH5.1 on a friends brand new
P2 450mhz 128M 8m scsi, when any user is logged in
on the console/virt terminals WHO lists everyone ok.
But if anyone is logged in via direct connection on
ttyS0 (mgetty process) then WHO will list other users
on the console/virt terminals and then hang, not showing
the user on the serial port.
Anyone have any theorys?
Anyone have any fixes?
--
The wealth of reality, cannot be seen from your locality.
------------------------------
From: bp jendrissek <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.hardware
Subject: Re: Pentium III Boycott and survey info
Date: 8 Mar 1999 13:44:30 GMT
BTW isn't this *slightly* off-topic? Doesn't this belong in
comp.lang.asm.x86?
In comp.os.linux.advocacy Anthony D. Tribelli <[EMAIL PROTECTED]> wrote:
: mlw ([EMAIL PROTECTED]) wrote:
: : Anthony D. Tribelli wrote:
: : > I don't think there was a reset instruction, documented or otherwise. I
Actually, there were 2 (loadall286 and loadall386) that could load values
into every (methinks) register visible in any other way (excluding electron
microscope, of course). The instruction took the values from 0:800h AFAIK.
Have a look at Ralf Brown's interrupt list - there are many similar
undocumented instructions. The problem with loadall (0fh 05h on '286;
0fh 07h on '386 IIRC) was that it was implemented only on *intel* 80286's.
Harris chips - no. Siemens, AMD - I don't know. Oh, and I think loadall
was not a protected instruction - anyone could execute it. Pretty stupid
leaving it there after development!
: : > believe the keyboard microcontroller was asked to reset the main CPU, and
: : > BIOS could recongnize a cold or warm boot and possibly jump to a location
: : > specified in RAM (to resume where things left off rather than FFFF:FFF0).
: : > To expand on your brief mention of 'kernel space', a protected mode OS
: : > (WinNT and Linux, maybe Win9x) can prevent user programs from doing this
: : > sort of thing.
Yes, also true, and more portable. Worked on *all* PC's, not just those
with "intel inside" (when was that phrase coined?)
[snip]
: : ... There was an undocumented instruction that is not in most
: : assemblers, but can be coded with db or emit. The instruction was put on
: : the chip so test program written by Intel could put the processor into
: : protected mode and take it back out again.
: This sounds like someone misunderstood an improved reset method which
: involved I/O ports (again prevantable if the OS chooses). This was
: supposedly an alternative the the much slower keyboard reset which OS/2
: did use. The faster method was not universally available.
No misunderstanding, loadall existed, 8042 trick always (?) worked, other
port gymnastics existed on some motherboards.
: I think yet another method involved causing multiple processor faults, and
: again an end user program could not do this.
This method depended (depends?) on circuitry that detects the cpu going
into shutdown - which only NMI or RESET can break IIRC. I think what
happens is that an exception is made to occur inside a double fault
(which can still be handled semi-gracefully) whereupon the cpu sends a
bus cycle telling the world it is tired of living.
: : It is this instruction that Microsoft used to enable its DOS box in OS/2
: : 1.x. I'm pretty sure it is a protected instruction, so a program would
: : have to be in an unprotected environment, such as Windows 9x or kernel
: : space in NT. The problem with the instruction was that it clobbered some
: : range of memory, I think 40H.
: I'm still highly skeptical of such an instruction existing. I'd love to
: see a URL, I suspect info got 'mutated' as it passed from one person to
: the next.
I doubt entries in Ralf Brown's list count as "mutated" - but then, who am
I to decide? Maybe I'll just fire up the old (Siemens cpu) 286 and see if I
can get loadall not to #UD.
Bernd Jendrissek
------------------------------
From: [EMAIL PROTECTED] (J.H.M. Dassen (Ray))
Subject: Re: crypt() linking error
Date: 8 Mar 1999 12:34:33 GMT
Kim Thye Chua <[EMAIL PROTECTED]> wrote:
>I am writing a program that needs to use crypt(const char *key, const char
>*salt) which should be a standard function in Linux. However I got a
>linking error after I compiled it. I am using RedHat 5.1, I guess RedHat
>uses another function for password checking.
No. It uses crypt() alright. But crypt() is in a separate library libcrypt,
so you need to add '-lcrypt' to your linker invocation.
HTH,
Ray
--
Obsig: developing a new sig
------------------------------
From: [EMAIL PROTECTED] (Rod Smith)
Subject: Re: Linux Safeguards for the Consumer?
Date: 8 Mar 1999 13:58:45 GMT
Reply-To: [EMAIL PROTECTED]
[Posted and mailed]
In article <01be6960$be9ba1c0$8a97d6d1@sher07>,
"Benjamin Sher" <[EMAIL PROTECTED]> writes:
> Dear friends:
>
> As a consumer with some experience with Dos, Win3.1 and Win95, I am in the
> process of deciding whether I should switch over to Linux. I am considering
> buying Linux on a Disk (LOAD) made by Cosmos Engineering
> (www.cosmoseng.com). I have been reading deeply in Red Hat Linux Unleashed
> in order to help me understand Linux and make a wise decision.
>
> Specs: NEC Pentium 166, MMX, 64 meg Ram. Win95 + Linux (hopefully) using
> LILO.
>
> QUESTIONS:
>
> 1 -- ROOT MODE SAFEGUARDS:
>
> Are there any safeguards (either command-line or by way of the GUI) against
> accidentally hitting the wrong key combination while in "root" mode?
No software can stop your fingers from doing something that's "wrong"
(unless you're a cyborg, of course, but that seems unlikely at this stage
of human technological development). Your main safeguard against doing
something destructive in root mode is to use the root account as little
as possible. You can also set up various protections to reduce the
likelihood of an incorrect command doing damage by altering the default
behavior of commands like rm to prompt you before doing anything. You do
this by creating an alias in your /root/.bashrc (and ~/.bashrc if you
want similar protections against damaging your user files as a user)
[this assumes you're running bash as your shell; it'll be different if
you use something else].
I'd point out that you can also trash a Windows system pretty badly by
deleting the wrong files. I don't know if Linux makes it easier or harder
to do this than Windows (I can think of reasons why it MIGHT be either
way).
> Does
> hitting such a combination of keys really result in destruction of your
> system (as opposed to "merely" requiring a reinstallation of Linux?
"Destruction of system" can mean either a total software wipout or
physical damage to the computer. In the case of stuff you can type as
root in Linux, the former is entirely possible (e.g., "rm -r /" will do a
pretty good job of wiping everything from your hard disk), but the latter
is unlikely. The former would, as you say, "merely" require software
re-installation. (Note that if you mount your Windows partition, such an
accident could wipe it out, too.)
> 2 -- SHUTDOWN SAFEGUARDS:
>
> I am planning to buy a battery to protect me from power failures, etc.
> Makes a lot of sense. But what about human error? Inadvertently hitting the
> power button to shut down the system. I understand that fsck or File System
> Checker is the LInux equivalent of Scan Disk. Is it a sufficient safeguard
> against human error?
Depends on what you mean by "safeguard." It certainly won't prevent you
from hitting the power button accidentally, but it'll check the system on
bootup the same as CHKDSK or ScanDisk would. Whether you're in Linux or
Windows, you'll PROBABLY be OK after such an accident, in terms of overall
filesystem integrity, but in both OSes there's a chance that you'll have
serious filesystem damage after a system crash or power outage.
> I have read that Linux, unlike Windows, runs in "unprotected" mode. What
> exactly does that mean, from a practical point of view?
I've no idea, without the exact context for this statement. Quite
frankly, it makes little sense to me as-is.
> Another way to put is: Is Linux ready for the consumer?
That depends on the consumer. Given the gist of your questions, I'll say
this: **NO** operating system in existence can protect people from their
own mistakes. Software design (and hardware design) can make it harder to
make mistakes, but if you're determined or clumsy enough, you can hose
just about anything. Linux goes about protecting itself against such
mistakes differently than does Windows. Linux uses security inherent in
its user accounts to keep ordinary users from trashing a system via
commands (though the power switch -- or an axe -- can be used to
circumvent this security). Windows seems to rely more upon people just
not TRYING to, say, delete the C:\WINDOWS directory (in truth, I've never
tried it, so I don't know what would happen if I did).
--
Rod Smith
[EMAIL PROTECTED]
http://www.channel1.com/users/rodsmith
NOTE: Remove the "uce" word from my address to mail me
------------------------------
From: Jean-Yves TOUMIT <[EMAIL PROTECTED]>
Subject: Problem with shadow - redhat 5.2 - kernel 2.2
Date: Mon, 08 Mar 1999 14:06:28 +0000
Hi all again ;-)
I downloaded and installed the shadow password suite. Seems to work (it
translated the password file and made a shadow file) but there is an
annoying message :
configuration error - unknown item 'CREATE_HOME' (notify administrator)
and as I'm the administrator of the machine, it won't help! ;-)
And more, when I su, the default PATH is /bin:/usr/bin which is rather
annoying in a su session!
I have RedHat 5.2, kernel 2.2.1.
If anybody has the same problem and solved it, I would be glad to hear
from you! ;-)
Thanks!
--
Jean-Yves TOUMIT
mailto:[EMAIL PROTECTED]
http://rfv-pc28.insa-lyon.fr
------------------------------
** 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 (and comp.os.linux.misc) via:
Internet: [EMAIL PROTECTED]
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-Misc Digest
******************************
