Sign in to follow this  
Followers 0
tony177

Boot error

58 posts in this topic

Apart from my lack of confidence in the effectiveness of things like swapping cables (and various other internal machinations) I am not motivated to do it at present before exploring sw related solutions since I have a medical problem with a (life-long) bad back and the logistics of getting at the computer to open it up frequently results in severe pain followed by a trip to the osteopath (once I am able to stand upright again) for treatment. Your comments about the various things that can deteriorate over time inside the computer are wholly valid but I still fall back on (or maybe am becoming fixated by) the unlikelihood of this being the problem bearing in mind the tight coupling of a normal boot failure faithfully followed by a CD boot success. Of course if that sequence breaks then it will be rethink time for me.

I think the best option in your case probably is sending your computer to repair shop for checking, including hardware checking due to your medical condition.

My previous motherboard is having booting problem intermitenly. Turn out some capacitor corrodes and leaked. Hard to see the corrosion even opening up the cover. In the end, I sent it to repair shop to check and replaced a new motherboard. That motherboard last me about 4 years before it start giving me problem.

Here is an update for which any comments from this haven of helpers would be welcome. For last two days system booted normally but then today back to boot failure. It seems to me something within my system is changing during normal operation or shutdown so in the spirit of the scientific method I took the following action immediately upon boot failure to explore the nature of the problem:

1) booted with spinrite which could see both disks and the two partitions in primary disk

2) booted with ntfs4dos which could run chksk and displayed a list of disks and partitions but didn't recognize the disks when I tried to navigate to them ('invalid drive specification)

3) booted again with normal windows start up but boot failed

4) booted with Windows CD into repair mode and navigated to all disks but did nothing else

5) booted again with normal windows start up and boot successful

That has been the scenario from the start - intermittent boot failure followed by booting with Windows CD enables subsequent successful normal boot. So, do any of you rodeo regulars interpret that general history supplemented with the specific sequence of actions and outcomes above as being indicative of or consistent with a hw problem? If so what hw problem specifically and if not, any idea what kind of other specific problem might be indicated.

0

Share this post


Link to post
Share on other sites

That has been the scenario from the start - intermittent boot failure followed by booting with Windows CD enables subsequent successful normal boot. So, do any of you rodeo regulars interpret that general history supplemented with the specific sequence of actions and outcomes above as being indicative of or consistent with a hw problem? If so what hw problem specifically and if not, any idea what kind of other specific problem might be indicated.

Is the Windows CD "loud" when loading? (i.e. a lot of noise and vibration from the CD drive when it seeks files)?

If yes, it may be still:

Anyone have any ideas as to what would intermittently cause the computer to behave like this and why the problem would be fixed after booting from the Windows CD in repair mode (does that replace some file(s) on the hard drive?).

I have seen defective hard disk cables/connectors behave like that.

but if you wish, I would throw in - without additional costs - also a cold solder joint anywhere on the motherboard or HD PCB.

+1. In other words: we here work by the Scientific Method. We construct hypotheses (not necessarily always explicitly described) and propose experiments in order to validate or disprove them. Then the OP performs the tests suggested and reports back. Then, more refined hypotheses are thought up (not necessarily by the same person) and so on, until we get to understand what actually is happening, and then we move on to solution. Without experiments we simply cannot help.

Naaah.

That's the "normal" approach. :thumbup

You well know how In some cases - when the post date+time sums up to an even number bigger than 6 :w00t: - I draw randomly a new thread and in it, instead of posting the usual good advice, I simply throw on the table a few (sometimes wild, sometimes educated) semi-random guesses :ph34r: , and then let the OP choose among them which ones he/she feels better with or is more convenient to him/her.

Other times I simply post real tall stories and see if the newcomer "buys" them :angel .

Once upon a time there was a PC that could not boot from it's internal hard disk.

This happened some times, and some other times the PC booted normally.

The queer thing was that just by inserting in the CD-ROM drive an XP install disc and bootng from it (but not doing any repair - or more generally operation from it) the next time the PC would boot normally.

When the PC doesnt boot it shows however a message "disk read error occurred", but right after a boot CD was just loaded it started booting normally again.

After some time of tests (new tests, not always the same ones) it was found how the vibrations induced to the case by booting from a CD made a false contact somewhere in the hardware "behave" (temporarily).

The actual cause was never found, nor the problem solved, as the PC's owner found it not convenient (due to a bad back medical condition) to actually doing the only logical thing, i.e. to open the case (and while at it cleaning it's internals), and try re-seating cables/cards/RAM etc. and visually inspect the parts before re-assembling...

both the PC and it's owner nonetheless lived happily ever after, with the only issue that the bad back problem became worse and worse each time the user had to bend to insert a boot CD to revive the stoopid PC....

jaclaz

0

Share this post


Link to post
Share on other sites

Once upon a time there was a PC that could not boot from it's internal hard disk.

This happened some times, and some other times the PC booted normally.

The queer thing was that just by inserting in the CD-ROM drive an XP install disc and bootng from it (but not doing any repair - or more generally operation from it) the next time the PC would boot normally.

When the PC doesnt boot it shows however a message "disk read error occurred", but right after a boot CD was just loaded it started booting normally again.

After some time of tests (new tests, not always the same ones) it was found how the vibrations induced to the case by booting from a CD made a false contact somewhere in the hardware "behave" (temporarily).

The actual cause was never found, nor the problem solved, as the PC's owner found it not convenient (due to a bad back medical condition) to actually doing the only logical thing, i.e. to open the case (and while at it cleaning it's internals), and try re-seating cables/cards/RAM etc. and visually inspect the parts before re-assembling...

both the PC and it's owner nonetheless lived happily ever after, with the only issue that the bad back problem became worse and worse each time the user had to bend to insert a boot CD to revive the stoopid PC....

jaclaz

There is no strange noise or vibration at all when booting from the Windows CD. Next time the problem occurs and still paying homage to the scientific method, I will immediately boot from a linux cd followed by a normal boot to see if that works. I acknowledge the CD drive may be unleashing some exquisitely tuned unnoticeable vibration that without fail (so far) fixes the furtive hardware fault even though, as I pointed out, prior to booting with the Windows CD the disks were seen by spinrite booting from floppy and by NTFS4DOS from floppy (running chkdsk but not recognizing the disks when I tried to navigate to them - what can this mean?). If a Linux boot actually works I will then try 'booting' with some non-operating system cd to determine if any cd drive activity mends this elusive hardware fault. If that works as well, I shall return here to congratulate and thank you.

But until such time and without insisting I still find the cause of this consistent pattern of behavior hard to credit to defective hardware. However in light of the 'once upon a time' fable, I mean story, even if my scientific method honoring alternative CD tests fail to induce boot I may try giving the pc a personal manly shaking when booting normally and see if that works. But then as the pc wasn't perceptibly shaken during the 2 floppy boots when the disks could be detected prior to cd boot, I'm not sure those vibrations would be good enough to coax boot success.

0

Share this post


Link to post
Share on other sites

Has it EVER (YELLING) occurred to you back up the MBR sector and ATTACH IT? (Sooooo easy to do...) Inspection of it would/may "eliminate" that from the equation.

You had ALSO (yelling again) mentioned a partitioning tool but haven't indicated what it is.

SpinRite - Gibson Research (GRC) - we KNOW of that site and the controversial "ShieldsUp" test (of which I too occasionally use). Have you, by chance, tried alternatives (like the Manufacturer's) or any number of other FREE HDD test softwares (again - SEARCH button is the little "Gear" in the upper right corner of this very screen)?

Seriously, I find it quite impossible that a "random boot error" would occur against the HDD then booting from CD then booting to HDD ("it works!") error-free to be normal by any means. Again, SERIOUSLY (YELLING again) this condition+"fix" is unheard of (read as "impossible") unless something else is wrong.

P.S. "Backing" the MBR in no way alters it AS IT WOULD CURRENTLY EXIST!

1 - (going) Bad Disk

2 - Geometry incorrectly detected(?)

3 - Bad MBR (some kind of RootKit?)

4 - BIOS "randomly" changing Geometry parameters

5 - You have a "super-hidden" partition (could also check that with MBR sector upload)

Process of elimination would really help... Only a Parent says to a child "Because"... We are not children. B) ;):lol:

0

Share this post


Link to post
Share on other sites

Has it EVER (YELLING) occurred to you back up the MBR sector and ATTACH IT? (Sooooo easy to do...) Inspection of it would/may "eliminate" that from the equation.

You had ALSO (yelling again) mentioned a partitioning tool but haven't indicated what it is.

SpinRite - Gibson Research (GRC) - we KNOW of that site and the controversial "ShieldsUp" test (of which I too occasionally use). Have you, by chance, tried alternatives (like the Manufacturer's) or any number of other FREE HDD test softwares (again - SEARCH button is the little "Gear" in the upper right corner of this very screen)?

Seriously, I find it quite impossible that a "random boot error" would occur against the HDD then booting from CD then booting to HDD ("it works!") error-free to be normal by any means. Again, SERIOUSLY (YELLING again) this condition+"fix" is unheard of (read as "impossible") unless something else is wrong.

P.S. "Backing" the MBR in no way alters it AS IT WOULD CURRENTLY EXIST!

1 - (going) Bad Disk

2 - Geometry incorrectly detected(?)

3 - Bad MBR (some kind of RootKit?)

4 - BIOS "randomly" changing Geometry parameters

5 - You have a "super-hidden" partition (could also check that with MBR sector upload)

Process of elimination would really help... Only a Parent says to a child "Because"... We are not children. B) ;):lol:

- Ok, attached is mbr backup I took a couple of days ago.

-I stated a few days ago that the pc came partitioned and hence I don't know what partitioning tool was used

-GRC shields up has nothing to do with spinrite. I have run spinrite (L2 twice and L4 once), chkdsk, the manufacturers' tool (albeit in 'fast' check mode) and a number of disk utilities all of which have given the disk a clean bill of health

- I agree the problem's manifestation and resolution is not normal which uncoincidentally is the very reason I came to a specialist board. Notwithstanding the group-think verdict it must be a hw fault, I am of the opinion (but open to offers) that the root problem is sw in nature with a hw symptom but a sw solution. My guess is some manner of sw corruption sporadically taking place for some unknown reason that somehow gets addressed by booting from Windows CD (pretty straightforward....)

- I am confident the disk is healthy and not going bad

- no idea about its geometry (although I have the paragon alignment tool but never used it)

- I'll run a rootkit scan but would astounded (not to mention crushingly shocked and disappointed) if I were infected

- I can't comment on BIOS but I haven't made any changes to it for years. When I boot to the Windows CD I go into bios to select the CD drive and all the drives are showing in their correct order

- after a previous mention of hidden sector I ran a partitioning tool (not necessarily the one used to partition this computer) and it just showed the usual drives with no hidden partition

mbrCopy.zip

0

Share this post


Link to post
Share on other sites

P.S. "Backing" the MBR in no way alters it AS IT WOULD CURRENTLY EXIST!

Not really. :w00t:

Just for the record, once you try backing up a "Never run a NT OS" PC MBR through booting a NT based OS or PE, unless some "special" care is taken, the MBR will change (the Disk Signature is added to it).

@tony177

Most probably the issue with your MBR DATA is that you have an Extended partition in the LBA only space marked as 05 (CHS mapped) whilst it should really be 0F (LBA mapped), see:

http://www.win.tue.nl/~aeb/partitions/partition_types-1.html

this is not a real problem until you run NT based systems only (from 2K onwards) but it may be an issue if using NT 4.0 and some versions of MS-DOS (or more generally DOS).

jaclaz

0

Share this post


Link to post
Share on other sites
Just for the record, once you try backing up a "Never run a NT OS" PC MBR through booting a NT based OS or PE, unless some "special" care is taken, the MBR will change (the Disk Signature is added to it).
Stipulation noted... ;)

edit - sounds like a "partitioning" tool was used...

Edited by submix8c
0

Share this post


Link to post
Share on other sites
Most probably the issue with your MBR DATA is that you have an Extended partition in the LBA only space marked as 05 (CHS mapped) whilst it should really be 0F (LBA mapped), see:

http://www.win.tue.nl/~aeb/partitions/partition_types-1.html

this is not a real problem until you run NT based systems only (from 2K onwards) but it may be an issue if using NT 4.0 and some versions of MS-DOS (or more generally DOS).

@jaclaz, I'm out of my element here, but because of what you mentioned, could that be somehow temporarily "fixed" by BIOS changes to the boot order? I say this because tony177 implies he makes a BIOS change in order to boot to the Windows CD:

... When I boot to the Windows CD I go into bios to select the CD drive and all the drives are showing in their correct order ...

and then I assume he makes a BIOS change again to boot back to the HDD. I also assume that no BIOS change was needed to boot to the two floppies he mentioned which could be why trying them didn't "do" anything. But again I'm WAY out of my element here and just grasping at straws. Anyway, would you please point tony177 to a method he could use to correct the issue you noticed to see if that fixes hie problem? Of course, as I'm sure he will say, this doesn't really explain why his system will boot "normally" for several days between hiccups, does it?

Cheers and Regards

0

Share this post


Link to post
Share on other sites

@jaclaz, I'm out of my element here, but because of what you mentioned, could that be somehow temporarily "fixed" by BIOS changes to the boot order? I say this because tony177 implies he makes a BIOS change in order to boot to the Windows CD:

... When I boot to the Windows CD I go into bios to select the CD drive and all the drives are showing in their correct order ...

and then I assume he makes a BIOS change again to boot back to the HDD. I also assume that no BIOS change was needed to boot to the two floppies he mentioned which could be why trying them didn't "do" anything. But again I'm WAY out of my element here and just grasping at straws. Anyway, would you please point tony177 to a method he could use to correct the issue you noticed to see if that fixes hie problem? Of course, as I'm sure he will say, this doesn't really explain why his system will boot "normally" for several days between hiccups, does it?

Just to clarify, I don't make any changes to BIOS at any stage. When booting to the CD drive, I just open BIOS to select the CD Drive from the list and hit enter - no change to it. Both floppy boots accessed the disk to run diags on it (spinrite on one, chkdsk on the other) but the disks are not recognized if I try to navigate to them. Also, my system booted normally (at least once!) each day for about 4 years until this problem appeared a short while ago.

0

Share this post


Link to post
Share on other sites
Notwithstanding the group-think verdict it must be a hw fault, I am of the opinion (but open to offers) that the root problem is sw in nature with a hw symptom but a sw solution. My guess is some manner of sw corruption sporadically taking place for some unknown reason that somehow gets addressed by booting from Windows CD (pretty straightforward....)

No. There's a serious misunderstanding here. There's no group-think verdict at all.

Some of us hazarded guesses about what might be happening. I didn't hazard any and continue not doing it.

For me, there's still too little info to go on. I'll reserve any guessing on my part for a later time, when I think some guess might be warranted.

That said, and since none of the others seem familiar with SpinRite, I must add SpinRite 6 rocks! It is able to do some types of repair no other tool I know of can do (in particular it can sometimes "revive" floppies no other software can save, for time enough for a good image to be acquired from them).

But SpinRite 6 is only useful for a limited number of conditions, most of them hardware issues. And, of course, it may be used simply to detect and access disks. Are you using SpinRite 6, perchance? All earlier versions of SpinRite are too out-of-date to be useful for disks using Win 95 OSR2 - Win 7 (read this and this).

BTW, in what regards disks and disk hardware, I consider Steve Gibson an authority, but for all other matters, especially security and the internet, his opinions are somewhat controversial and should be taken with the proverbial grain of salt.

0

Share this post


Link to post
Share on other sites

Anyway, would you please point tony177 to a method he could use to correct the issue you noticed to see if that fixes hie problem? Of course, as I'm sure he will say, this doesn't really explain why his system will boot "normally" for several days between hiccups, does it?

No, it is a "marginal issue".

The "traditional" Partition ID for the Extended Partition has always been 05.

The 0F - which means "05 but LBA" - was introduced to prevent OS's that use CHS from accessing it, just like 0C and 0E.

From a NT OS viewpoint a partiion that occupies LBA only accessible space but that is marked CHS doesn't make any "real" difference, as NT will use LBA anyway.

Due to the way the LBA only space is mapped in CHS, you have a number of sectors (from 1023/0/1 up to 1023/254/63 that are accessible through CHS) see tony177's PC partition table:

Entry Type Boot bCyl bHead bSect eCyl eHead eSec StartSector NumSectors

#0 07 80 0 1 1 1023 254 63 63 436919742

#1 05 00 1023 0 1 1023 254 63 436919805 813338820

For the record, OT but JFYI, newer XP SP and Vista :ph34r:, and later do map them when partitioning as 1023/254/63 up to 1023/254/63, and this is an additional "prevention measure", though there is a serious "quirk" when using XP Disk Management on a Vista :ph34r: or 7 created disk:

http://reboot.pro/9897/

Traditionally (though actually "wrongly") partition ID's have been intended as "reliable info of filesystems", while their actual use is that of preventing access/mounting from "legacy" Operating Systems, see:

http://homepage.ntlworld.com./jonathan.deboynepollard/FGA/determining-filesystem-type.html

http://www.forensicfocus.com/Forums/viewtopic/t=8880/

A number of partitioning tools (or however MBR editing ones) may use 05 improperly, but until you run on that disk an OS or tool that interprets the 05 "literally" there won't be problems.

To "fix" it the only thing needed is to change byte 0x1D2 from current value 0x05 to 0x0F.

jaclaz

0

Share this post


Link to post
Share on other sites
Notwithstanding the group-think verdict it must be a hw fault, I am of the opinion (but open to offers) that the root problem is sw in nature with a hw symptom but a sw solution. My guess is some manner of sw corruption sporadically taking place for some unknown reason that somehow gets addressed by booting from Windows CD (pretty straightforward....)

No. There's a serious misunderstanding here. There's no group-think verdict at all.

Some of us hazarded guesses about what might be happening. I didn't hazard any and continue not doing it.

For me, there's still too little info to go on. I'll reserve any guessing on my part for a later time, when I think some guess might be warranted.

That said, and since none of the others seem familiar with SpinRite, I must add SpinRite 6 rocks! It is able to do some types of repair no other tool I know of can do (in particular it can sometimes "revive" floppies no other software can save, for time enough for a good image to be acquired from them).

But SpinRite 6 is only useful for a limited number of conditions, most of them hardware issues. And, of course, it may be used simply to detect and access disks. Are you using SpinRite 6, perchance? All earlier versions of SpinRite are too out-of-date to be useful for disks using Win 95 OSR2 - Win 7 (read this and this).

BTW, in what regards disks and disk hardware, I consider Steve Gibson an authority, but for all other matters, especially security and the internet, his opinions are somewhat controversial and should be taken with the proverbial grain of salt.

The impression I get/got (at least at the outset) was that 'everyone' who commented deemed the problem to be due some manner of defective hardware, which of course it might, but that's not my opinion for the reasons I've stated. I am using spinrite 6 which, in common with all the other disk utilities I've run, has given the disk a clean bill of health - no problems identified at all. I am quite familiar with Steve Gibson's stuff and some of the controversy he has provoked about security but I have never encountered any negative comments about spinrite which seems to be very highly regarded as a disk utility.

0

Share this post


Link to post
Share on other sites

if you have other spare box with same hardware configuration,

you might want to move the HDD into that box, and see if the problem also occured there.

0

Share this post


Link to post
Share on other sites

if you have other spare box with same hardware configuration,

you might want to move the HDD into that box, and see if the problem also occured there.

Indeed that would be an informative test but I don't have a spare box.

0

Share this post


Link to post
Share on other sites

Hmmmm. I just noted something - What type of PC is this? (I've gone through the thread and if info was there I missed it?)

Hardware errors will occur if it's "nasty" inside. Any number of things can go wrong, including the CPU overheating. You DID say "4 years and then it fails"... A can of "dry air" and cracking it open, blowing it out, and closing it up (check the cables while you're at it) might "fix" it...

P.S. "lappies" (afaik) may be more prone to being "dirty" since there is way too little "space" inside...

Every indication (even searching the internet) points to some kind of intermittent hardware problem.

0

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0

  • Recently Browsing   0 members

    No registered users viewing this page.