RAIDING different size drives

[Rod Speed]

David Brown wrote

Rod Speed wrote

Eventually there's a limit to how long I'm going to continue this thread.

Thats true of any thread.

I know you have a lot of experience and knowledge, and there's plenty that I can benefit from (the tip about Barts
being an example). But I'm fed up trying to apologise because /you/ can't understand sarcasm and exaggeration

That is neither. Its just sillyness.

(I'm British - it's part of our culture and language), and I'm fed up with you repeatedly changing the discussion so
you can cut-and-paste "you're wrong" at every paragraph.

I changed nothing there.

You just hate having your nose rubbed in your terminal stupiditys.

Telling me I'm foolish for my personal preferences of how I use a computer

Didnt do that. Thats a particular phraseology that even a stupid pom should understand.

And whatever you prefer, you cant justify your preferences with logic.

The problem with separate partitions is that its difficult to get the
partition size right, particularly as the system is used over time and
very dangerous to resize partitions without a full image of the entire
physical drive, and that often isnt feasible with home systems.

And you dont gain a damned thing with separate partitions anyway.

is the last straw.

Your problem.

Maybe you think yourself so high-and-mighty and infallible that you'll plonk me for this.

I dont ever plonk anyone, ever.

The most I ever do is ignore fools, like the one I chose to ignore
today who maintains that the CIA did 9/11 and not some muslims.

If not, perhaps we'll meet again on another threat and get off on a better foot.

Unlikely. I put the boot into sillier ideas when I come across them.

You get to like that or lump it.

mvh.,

gte

 

[John Turco]

David Brown wrote:

Another thing to consider is your access to the files in the case of
operating system death, which is not exactly uncommon on desktop windows
systems. In such cases, your data is safe on the disk, but the windows
registry or critical files are corrupted. Windows will do this on its
own occasionally, especially if provoked (such as by power failures),
and third-party "security" software updates are notorious for rendering
windows unbootable. And of course, malware of all sorts can similarly
render the machine useless.

How do you then get your data files off the broken system, before trying
to fix it or to use a "system restore" CD that deletes all your data?
Ideally, you use the copies that are on a server, or your backup copies.
If you've got a second copy on an independent hard disk, you can use
that disk in another machine, or remove it until you've fixed the main
drive. You can also use a live Linux CD and access the files that way.
But if everything is on a raid setup, you don't have an independent
way to access the data. Maybe a live Linux CD will be able to access
the data, maybe not.

Hello, David:

This "operating system death" happened to me, twice (on my Windows
XP box), within a roughly 8-month period. My only solution, each
time, was to swap around my two Samsung 160GB internal hard disks,
and reinstall XP and my application software, on the drive which
had previously been "D:."

The old C: became the new D:, and I could see and access D:, from
the new C: -- and also, copy over whatever little data that I'd
earlier failed to back up (to my "SimpleTech" external 160GB HDD).

Confused, yet? <g> Seriously, what causes such a problem, and can
it be prevented?

I'm considering buying and installing a SATA controller card, to
replace the PATA card, on my ancient AT computer (Pentium III).
Then, I'd install a pair of Samsung 500GB SATA puppies (which had
been intended for an upcoming PC build).

As I'm beginning to run out of space, on my present 160GB drive,
I needed to upgrade, anyway (and at a later date, I'll purchase
a couple of Samsung 1TB models, to put into my "dream machine"
of the future).

Do you think that my plan would eliminate the possibility of
"operating system death?" I simply bring this up, because I have
a suspicion (however unfounded) my PATA card may be the culprit.

Thanks, in advance!


Cordially,
John Turco <[email protected]>


PS: In another post, you mentioned that you're in Norway (and
I noticed the word "Sweden," in your message headers). In any
event, "David Brown" doesn't sound very Scandinavian, and your
English is quite excellent.

 

[Rod Speed]

John Turco wrote

David Brown wrote

This "operating system death" happened to me, twice
(on my Windows XP box), within a roughly 8-month period.

What were the symptoms of the 'death' ?

My only solution, each time, was to swap around my two
Samsung 160GB internal hard disks, and reinstall XP and my
application software, on the drive which had previously been "D:."

Did you try a repair install ?

The old C: became the new D:, and I could see and access D:,
from the new C: -- and also, copy over whatever little data that I'd
earlier failed to back up (to my "SimpleTech" external 160GB HDD).

Confused, yet? <g>

Seriously, what causes such a problem,

Depends on the symptom you were getting with the 'death'

and can it be prevented?

Yes, something must have caused that since thats a very
uncommon result and whatever caused it can be fixed.

I'm considering buying and installing a SATA controller card, to
replace the PATA card, on my ancient AT computer (Pentium III).
Then, I'd install a pair of Samsung 500GB SATA puppies
(which had been intended for an upcoming PC build).

As I'm beginning to run out of space, on my present 160GB drive,
I needed to upgrade, anyway (and at a later date, I'll purchase
a couple of Samsung 1TB models, to put into my "dream machine"
of the future).

Do you think that my plan would eliminate the possibility of "operating system death?"

Depends on what caused the previous 'death'

If its a bad power supply or motherboard, no it wont.

I simply bring this up, because I have a suspicion
(however unfounded) my PATA card may be the culprit.

Much more likely to be a power supply or motherboard.

 

[David Brown]

David Brown wrote:




Hello, David:

This "operating system death" happened to me, twice (on my Windows
XP box), within a roughly 8-month period. My only solution, each
time, was to swap around my two Samsung 160GB internal hard disks,
and reinstall XP and my application software, on the drive which
had previously been "D:."

The old C: became the new D:, and I could see and access D:, from
the new C: -- and also, copy over whatever little data that I'd
earlier failed to back up (to my "SimpleTech" external 160GB HDD).

Confused, yet? <g> Seriously, what causes such a problem, and can
it be prevented?

I'm considering buying and installing a SATA controller card, to
replace the PATA card, on my ancient AT computer (Pentium III).
Then, I'd install a pair of Samsung 500GB SATA puppies (which had
been intended for an upcoming PC build).

As I'm beginning to run out of space, on my present 160GB drive,
I needed to upgrade, anyway (and at a later date, I'll purchase
a couple of Samsung 1TB models, to put into my "dream machine"
of the future).

Do you think that my plan would eliminate the possibility of
"operating system death?" I simply bring this up, because I have
a suspicion (however unfounded) my PATA card may be the culprit.

I doubt this will do anything significant to avoid the risk of
"operating system death". The main causes of an OS "dying" and been
unbootable, in no strict order, are:

1. Corrupt registry
2. Malware
3. Bug in the OS
4. Failed update of the OS
5. Conflict with security software
6. Lost passwords or encryption keys
7. Other user error (such as being too smart...)
8. Hard disk errors
9. Other hardware issues (such as RAM problems)

I might have missed some, and the risks of these will vary, but that's
what comes off the top of my head. You'll note that 1 through 5 are
almost exclusively windows issues.

You can mess up the configuration files of a Linux system, but your risk
is much lower than for the Windows registry.

Linux (and MacOS) are not completely immune to malware, but the risk is
hugely less.

You can get bugs in any OS, but the critical parts of the Linux kernel
and drivers are smaller and more modular than for windows, and you don't
have a large mass of third-party drivers installed at the kernel level
(third-party drivers are the cause of the the great majority of Windows
OS-level crashes).

Updates of a Linux system are less risky, because most distros will keep
previous kernels and give you the option of going back at boot time.
It's not impossible to get yourself in trouble with an update,
especially if it is a major distro version update, but it's hard. To be
extra safe, stick to the more "mature" and well-tested distros rather
than the bleeding edge types. Of course, most Windows updates go fine
too - we are talking about possible risks, not likely problems.

And of course on Linux you don't get the situation where Norton Internet
Security (or whatever) decides that ntkernel.exe is a virus and fails to
boot. (Though people have been known to use things like selinux to lock
themselves out of their Linux systems.)

One thing to note about the above is that it depends somewhat on how you
define an "operating system death". It's perfectly possible for a user
to corrupt their gnome or KDE user settings to the point where they
can't work with their desktop, or to have problems with the GUI on
Linux. You then have options like logging in as a different user, or
working from the command line so that you can fix the system. This is,
if you like, a sort of half-death state that you don't really get with
windows (the nearest windows can get is "works in safe mode only").
Whether you consider it as workable or effectively broken is a matter of
how comfortable you are with the lower levels of Linux.


Hardware problems are always an issue with any system. You reduce the
risks here by buying decent quality components (check reviews before you
buy, and avoid things that haven't been on the market long - no one knew
that "DeathStar" disks were bad when they first arrived). And of
course, given the thread, raid can help enormously for hard disk problems.


All in all, my recommendation is to get a small file server running
Linux (or BSD - the differences for a server are minor if you are not a
particular enthusiast). You can buy such systems ready-made. This will
give you a large and safe space to store all your files. I think this
would be much more sensible than piling even more onto your existing
system - you won't make it any more reliable by adding more hardware to
it, and it will make the transition to a new machine much smoother (all
your data remains in the same place).

Thanks, in advance!


Cordially,
John Turco <[email protected]>


PS: In another post, you mentioned that you're in Norway (and
I noticed the word "Sweden," in your message headers). In any
event, "David Brown" doesn't sound very Scandinavian, and your
English is quite excellent.

I'm Scottish, but live in Norway.

 

[Rod Speed]

David Brown wrote

John Turco wrote

I doubt this will do anything significant to avoid the risk of "operating system death".

You did get that right.

The main causes of an OS "dying" and been unbootable, in no strict order, are:

We'll see...

1. Corrupt registry

Hardly ever.

2. Malware
3. Bug in the OS

Hardly ever.

4. Failed update of the OS

Almost never.

5. Conflict with security software

Almost never.

6. Lost passwords or encryption keys

Mindlessly silly.

7. Other user error (such as being too smart...)

Even sillier.

8. Hard disk errors

Should have been at the top.

9. Other hardware issues (such as RAM problems)

Almost never with ram.

I might have missed some, and the risks of these will vary, but that's what comes off the top of my head.
Waffle.

You'll note that 1 through 5 are almost exclusively windows issues.

And hardly ever are a problem.

You can mess up the configuration files of a Linux system, but your risk is much lower than for the Windows registry.
Lie.

Linux (and MacOS) are not completely immune to malware, but the risk is hugely less.

Its completely trivial for anyone with a clue to avoid malware.

You can get bugs in any OS, but the critical parts of the Linux kernel and drivers are smaller and more modular than
for windows,

Irrelevant to the risk of bugs. They are also used by a hell of a lot less than win.

and you don't have a large mass of third-party drivers installed at the kernel level (third-party drivers are the
cause of the the great majority of Windows OS-level crashes).

Irrelevant to what makes a system unusable.

Updates of a Linux system are less risky, because most distros will keep previous kernels and give you the option of
going back at boot time.

Pity about restore points with Win.

It's not impossible to get yourself in trouble with an update,
especially if it is a major distro version update, but it's hard.

Just as true of Win.

To be extra safe, stick to the more "mature" and well-tested distros
rather than the bleeding edge types. Of course, most Windows updates
go fine too - we are talking about possible risks, not likely problems.

And you're ignoring restore points.

And of course on Linux you don't get the situation where Norton
Internet Security (or whatever) decides that ntkernel.exe is a virus
and fails to boot. (Though people have been known to use things like selinux to lock themselves out of their Linux
systems.)

One thing to note about the above is that it depends somewhat on how you define an "operating system death". It's
perfectly possible for a user to corrupt their gnome or KDE user settings to the point where they can't work with
their desktop, or to have problems with the GUI on Linux. You then have options like logging in as a different user,
or working from the command line so that you can fix the system.

Nothing like a repair install with Win.

This is, if you like, a sort of half-death state that you don't really get with windows
Wrong.

(the nearest windows can get is "works in safe mode only").
Wrong.

Whether you consider it as workable or effectively broken is a matter of how comfortable you are with the lower levels
of Linux.
Wrong.

Hardware problems are always an issue with any system. You reduce the risks here by buying decent quality components
Wrong.

(check reviews before you buy, and avoid things that haven't been on the market long - no one knew that "DeathStar"
disks were bad when they first arrived).

They took quite some time before their downsides showed up.

And of course, given the thread, raid can help enormously for hard disk problems.

All in all, my recommendation is to get a small file server running Linux

More fool you. Thats pointless for most home users.

(or BSD - the differences for a server are minor if you are not
a particular enthusiast). You can buy such systems ready-made.
Pointless.

This will give you a large and safe space to store all your files.

Makes no sense for most home systems.

I think this would be much more sensible than piling even more onto your existing system -

More fool you.

you won't make it any more reliable by adding more hardware to it,

Wrong, most obviously with mirroring.

and it will make the transition to a new machine much smoother (all your data remains in the same place).

And the new system is crippled by that stupid approach.

I'm Scottish, but live in Norway.

More fool you.

 

[David Brown]

David Brown wrote

I really don't know what world you live in. It certainly doesn't always
correspond to the rest of computer users (home or corporate users,
experts or amateurs).

I'm giving some links here, because I doubt if you'll credit any
personal experiences. They are mostly found with a quick google.

Hardly ever.

<http://support.microsoft.com/kb/307545>

etc., etc., etc.

Microsoft's website search reveals hundreds of other pages.

Hardly ever.

I'm guessing the "hardly ever" refers to "bug in OS", while you've
skipped the big one for windows - malware.

Google gives 45 million hits for "bug in windows". I'm sure with a
little effort you'll find the odd case of a bug in the OS itself (rather
than the bundled apps). MS occasionally provides updates and service
packs - sometimes these are to fix bugs.

Almost never.
<http://peter.vox.com/library/post/s...925902-update-left-my-machine-unbootable.html>


Almost never.

<http://www.dslreports.com/forum/r22522031-McAfee-Virus-update-leaves-PCs-unbootable>


Its completely trivial for anyone with a clue to avoid malware.

Here's a newsflash for you - most windows users *don't* have a clue. A
great many IT administrators running windows networks, even in large
installations, don't have a clue either.

Irrelevant to the risk of bugs. They are also used by a hell of a lot
less than win.

Do you know anything about software development for large projects?
Statistically speaking, the number of bugs per line of code is
surprisingly consistent over a range of projects, development models,
and programming languages. When a project is big enough, old enough,
and has enough developers with a spread of talents, more code means more
bugs.

Irrelevant to what makes a system unusable.

So when a windows system BSODs, it is only due to windows errors (or
hard disk failures, of course). Not all BSODs make a system unusable,
of course - but they do if they keep repeating.

Wrong.

Sorry, I admit I got this completely wrong. It's always been a well
known fact that buying the cheapest and lowest quality parts is the best
way to improve the reliability and lifetime of any system. People only
ever buy components from brands with established reputations, rather
than cheap copies, because they look cool.

More fool you.

You are now an expert in where I should choose to live, amongst all your
other talents? Is this based on your long personal experience with both
these countries, or at least your long research and accumulation of
knowledge in this area?

 

[Arno]

David Brown wrote:

<heavily edited for brevity>

Hello, David:

This "operating system death" happened to me, twice (on my Windows
XP box), within a roughly 8-month period. My only solution, each
time, was to swap around my two Samsung 160GB internal hard disks,
and reinstall XP and my application software, on the drive which
had previously been "D:."

I have had that also several times. My solution is to keep as little
as possible on c: and have a periodic image (sector-wise) backup of
c: (made with Linux, but other options exist). When XP chooses to die
again or behave erratically, it is usually fixed by reinstalling
the last snapshot on c:. This takes about 5 minutes. Once I had
to go back something like 8 weeks, but typically the last snapshot
(2-4 weeks) has worked fine. Drawback is that you have to reinstall
some of the software you installed in between.

It is also highly advisable to take a new snapshot before installing
new software.

Arno

 

[Rod Speed]

David Brown wrote

Rod Speed wrote

I really don't know what world you live in.

Your problem, as always.

It certainly doesn't always correspond to the rest of computer users (home or corporate users, experts or amateurs).

You're lying now.

I'm giving some links here, because I doubt if you'll credit any
personal experiences. They are mostly found with a quick google.

None of them are any news.

etc., etc., etc.

Microsoft's website search reveals hundreds of other pages.

Doesnt say a damned thing about how often that happens.

I'm guessing the "hardly ever" refers to "bug in OS", while you've skipped the big one for windows - malware.

You quite sure you aint one of those rocket scientist fellas ?

Google gives 45 million hits for "bug in windows".

Pity we happened to be discussing what often produces AN UNUSABLE OS INSTALL.

I'm sure with a little effort you'll find the odd case of a bug in the OS itself (rather than the bundled apps). MS
occasionally provides updates and service packs - sometimes these are to fix bugs.

Pity we happened to be discussing what often produces AN UNUSABLE OS INSTALL.

<http://peter.vox.com/library/post/s...925902-update-left-my-machine-unbootable.html>
<http://forums.techarena.in/windows-security/1163938.htm>

I didnt say never. The world ALMOST is there for a reason.

And it cant be the problem in John's case, because the new
clean install on the other drive would get the same result.

<http://www.dslreports.com/forum/r22522031-McAfee-Virus-update-leaves-PCs-unbootable>
<http://news.softpedia.com/news/AVG-Faulty-Update-Leaves-Computers-Unbootable-97766.shtml>

I didnt say never. The world ALMOST is there for a reason.

And it cant be the problem in John's case, because the new
clean install on the other drive would get the same result.

Here's a newsflash for you
Nope.

- most windows users *don't* have a clue.

Wrong, as always.

A great many IT administrators running windows networks, even in large installations, don't have a clue either.

Corse that never ever happens with other than Win, eh ?

Do you know anything about software development for large projects?

Yep, including OSs, unlike you.

Statistically speaking, the number of bugs per line of code is surprisingly consistent over a range of projects,
development models, and programming languages.

Irrelevant to the FACT that Win gets a hell of a lot more using
it than anything else does and so the warts show up sooner.

When a project is big enough, old enough, and has enough developers with a spread of talents, more code means more
bugs.

Irrelevant to the FACT that Win gets a hell of a lot more using
it than anything else does and so the warts show up sooner.

So when a windows system BSODs, it is only due to windows errors (or hard disk failures, of course).

Never ever said anything even remotely resembling anything like that.

I JUST said that those OS level crashes due to drivers are an
entirely separate matter to what is being discussed, what is the
likely reason that a particular OS INSTALL BECAME UNUSABLE.

Not all BSODs make a system unusable,
of course - but they do if they keep repeating.

And that isnt common.

Sorry, I admit I got this completely wrong. It's always been a well
known fact that buying the cheapest and lowest quality parts is the
best way to improve the reliability and lifetime of any system.

Never ever could bullshit its way out of a wet paper bag.

People only ever buy components from brands with established
reputations, rather than cheap copies, because they look cool.

Never ever could bullshit its way out of a wet paper bag.

You are now an expert in where I should choose to live, amongst all your other talents?

That was sarcasm, ****wit.

Is this based on your long personal experience with both these countries, or at least your long research and
accumulation of knowledge in this area?

You'll end up completely blind if you dont watch out, child.

 

[JK Jones]

I have had that also several times. My solution is to keep as little
as possible on c: and have a periodic image (sector-wise) backup of
c: (made with Linux, but other options exist). When XP chooses to die
again or behave erratically, it is usually fixed by reinstalling
the last snapshot on c:. This takes about 5 minutes. Once I had
to go back something like 8 weeks, but typically the last snapshot
(2-4 weeks) has worked fine. Drawback is that you have to reinstall
some of the software you installed in between.

It is also highly advisable to take a new snapshot before installing
new software.

So stupid that it hasnt even noticed what a repair install and restore points are about.

 

[John Turco]

Sydney Lambe wrote


That isnt a good idea, it prevents anyone from researching
the issues being discussed using groups.google.

<heavily edited for brevity>

Hello, Rod:

I wonder...does this "X-No-Archive: yes" statement, need to
appear in the message headers, in order to be valid?


Cordially,
John Turco <[email protected]>

 

[John Turco]

David Brown wrote:

Yes, email is the big issue. Everything else is typically saved as
simple files, and is easy to find and backup or copy to a new
installation or new PC. But email is often a problem to preserve, with
many people using Outlook Express and POP3. Their email history is
locked into a non-standard format mailbox, with important settings
scattered about the registry and other files. I'm not always convinced
by the privacy policies of online email systems (gmail et.al.), but they
are far preferable to POP3 (I prefer IMAP on a server, but IMAP servers
are rare in homes).

<edited>

Hello, David:

If you check my message headers, you'll notice that I'm still using
Netscape Communicator 4.8. This is the final version of the program
(released in 2003) and during the interim, Netscape, itself, has
gone out of business.

Why do I bother with such an old, unsupported application, you may
ask? Well, it's because I know how to back up its folders, to my
external hard drive. Then, whenever I must reinstall Windows and
everything else, I merely copy those folders over, afterward.

This method allows me to keep my e-mail and Usenet data completely
intact, and without any "seams." I'm unaware of how to do such a
thing, in other browsers/readers (and I absolutely despise Outlook
Express, besides).

Incidentally, I've been unable to send e-mail, with Communicator,
since September of 2006; I assume that my dial-up ISP (Concentric)
made some kind of change, then, which caused the problem.

Hence, I've needed to employ Outlook Express, for this purpose.
(Plus, I use the "Send Later" function, in Communicator, so that
all of those messages appear there, also.)

It's all rather kludgy, and yet, fairly effective!


Cordially,
John Turco <[email protected]>

 

[John Turco]

David Brown wrote


Didnt do that. Thats a particular phraseology that even a stupid pom should
understand.


gte

Hello, Rod:

Would you explain the meanings of following abbreviations, please?

pom

mvh

gte

Thanks!


Cordially,
John Turco <[email protected]>

 

[John Turco]

John Turco wrote


What were the symptoms of the 'death' ?

Hello, Rod:

My PC simply sat there, with a blinking, white cursor, at the bottom-left of
the monitor's black screen. On each occasion, it was when I first turned it
on, that day.

Did you try a repair install ?

Well, I attempted a "system repair," using my XP CD's built-in "Recovery
Console."

Depends on the symptom you were getting with the 'death'

True, but, there was only one symptom (i.e., the blinking cursor).

Yes, something must have caused that since thats a very
uncommon result and whatever caused it can be fixed.

That's what I'd thought, too.

Depends on what caused the previous 'death'

If its a bad power supply or motherboard, no it wont.


Much more likely to be a power supply or motherboard.

Of those two components, my own hunch would be the power supply (7-year
old Enermax). Still, it's always behaved itself, otherwise, and this "OS
death" phenomenon has been a rather rare occurrence, during my 14+ years
of computing. (From 1995 to September, 2008, it was entirely unheard of.)


Cordially,
John Turco <[email protected]>

 

[John Turco]

John Turco wrote:


I doubt this will do anything significant to avoid the risk of
"operating system death". The main causes of an OS "dying" and been
unbootable, in no strict order, are:

1. Corrupt registry
2. Malware
3. Bug in the OS
4. Failed update of the OS
5. Conflict with security software
6. Lost passwords or encryption keys
7. Other user error (such as being too smart...)
8. Hard disk errors
9. Other hardware issues (such as RAM problems)

Hello, David:

I believe that 2, 8, and 9 may be applicable, but, not the others.

<large cut of pro-Linux advice>

Sorry, man...I'm a Windows "lifer." <g> In fact, for my upcoming
home-built PC project, I'd already bought an OEM version of Vista
Home Premium 64-bit (with SP1), last year.

I'm Scottish, but live in Norway.

Those infamous Vikings invaded Scotland, many centuries ago; thus,
turnabout is fair play, no? <g>


Cordially,
John Turco <[email protected]>

 

[John Turco]

David Brown wrote


They took quite some time before their downsides showed up.

<edited>

Hello, Rod:

I want David Brown to understand that, "buying decent quality components" has always been my goal --
and one which I've largely accomplished, in my estimation.

For example, I first heard of what a great value Samsung hard disks are, a number of years ago, by
reading your recommendations of this particular brand, here in <as a result, I've been a satisfied Samsung (internal) HDD purchaser, exclusively, since 2004.


Cordially,
John Turco <[email protected]>

 

[John Turco]

Rod Speed wrote:


Sorry, I admit I got this completely wrong. It's always been a well
known fact that buying the cheapest and lowest quality parts is the best
way to improve the reliability and lifetime of any system. People only
ever buy components from brands with established reputations, rather
than cheap copies, because they look cool.


You are now an expert in where I should choose to live, amongst all your
other talents? Is this based on your long personal experience with both
these countries, or at least your long research and accumulation of
knowledge in this area?

Hello, David:

Rod Speed is Australian, for your information...and, during my 6+ years
as a regular reader of this newsgroup (and sometimes poster, to it), I've
found that he's very knowledgeable about computers.


Cordially,
John Turco <[email protected]>

 

[John Turco]

I have had that also several times. My solution is to keep as little
as possible on c: and have a periodic image (sector-wise) backup of
c: (made with Linux, but other options exist).

<edited>

Hello, Arno:

Well, that's basically what I've done, since being forced to start from
scratch, over a month ago. Leaving the vast majority of my data on D:,
will make it easier to recover from the next (potential) system disaster.

That is, instead of having to swap my two internal hard disks around, on
their respective IDE cables, I'll just reformat C: and proceed from there.

Although, after I replace my current 160GB drives with new 500GB models,
I'll likely return to my old configuration: OS, programs and data, on C:,
with D: being mostly reserved for audio/video capture, DVD "burning," etc.


Cordially,
John Turco <[email protected]>

 

[Rod Speed]

John Turco wrote

Rod Speed wrote

My PC simply sat there, with a blinking, white cursor, at the
bottom-left of the monitor's black screen. On each occasion,
it was when I first turned it on, that day.

OK. Most likely something hardware wise is corrupting the install.

Well, I attempted a "system repair," using my XP CD's built-in "Recovery Console."

You dont do a repair install that way.

You proceed as if you are doing a clean install AND DONT FORMAT
THE PARTITION YOU ARE INSTALLING TO. XP will say that its
found an existing install and will offer to repair it. You agree to that.

The big advantage over a clean install is that the settings and apps already installed arent affect.

True, but, there was only one symptom (i.e., the blinking cursor).

That usually means that the install has got corrupted.

If you use a logged install, you will usually see where its getting to when booting.

Since another clean install works, that does mean that you havent seen one
common cause of that symptom, some hardware dying and that causing the
device driver for that hardware not able to start that hardware at boot time.

That's what I'd thought, too.

Of those two components, my own hunch would be the power supply (7-year old Enermax).

Yeah, tho capacitor failure on the motherboard is quite common with that era hardware.

Still, it's always behaved itself, otherwise, and this "OS death" phenomenon
has been a rather rare occurrence, during my 14+ years of computing.
(From 1995 to September, 2008, it was entirely unheard of.)

Thats the problem with intermittent faults, they can be a complete pain in the arse.

 

[Rod Speed]

John Turco wrote

Rod Speed wrote

Hello, Rod:

Would you explain the meanings of following abbreviations, please?

pom

'person' from england. Otherwise known as whinging poms.

mvh
http://74.125.153.132/search?q=cach...opic=83063+mvh+means&cd=1&hl=en&ct=clnk&gl=au

gte

If I told you that, I'd have to kill you. Again.

Thanks!

Cordially,
John Turco <[email protected]>

Beerily,
Rod

 

[Rod Speed]

<heavily edited for brevity>

Hello, Rod:

I wonder...does this "X-No-Archive: yes" statement, need to
appear in the message headers, in order to be valid?

Nope, in fact it should appear in the body of the text of the post.
http://en.wikipedia.org/wiki/X-No-Archive