The box in question is a rather older machine, but has always been
steady and actually the one I use the most because it is out in my
workshop area. That said, it was originally win 7 that I installed win
10 on several years ago. It runs all day long usually, and I put it to
sleep at night. Not as fast as my newer one in my office, but it works
nicely with latest Firefox, BBird, Office 2019 and most other programs I
need.

So, a couple days ago, I would come to the computer and notice it had rebooted. The arrangement on my Core Temp taskbar icons were not as I
had left them, and the BBird was closed. So after a couple of these, I started investigating.

Event Viewer showed it booted because of a bugcheck error. This error
was the same for several reboots. It didn't reboot while I was using
it, and otherwise worked just fine.
So I did the DISM /restore which couldn't find a source. It did say the
files were repairable. So I got an win 10 iso and used it for the
source and it did say it repaired everything.
I then did an SFC/ scannow and it reported everything OK.
Next I did a chkdsk /f and it reported nothing to fix.

Used it for awhile and all seemed ok. Came back later and it had
rebooted again.

Checked on video driver and already had the best one.
Virus check and nothing there

Rebooted again, so I decided to try a system restore point. First time,
it rebooted after a short time doing the restore.

Did a memory check using the windows memory test. I know it's not the
best, but it said all was good.

Tried a second time to restore and after over 2 hours I finally shut it
down and restarted. Hated doing that, but it seemed to be stuck.

Decided to do a restore, saving my files, and it got to the first reboot
at about 12% and failed, resetting back to before the setup. I didn't
catch the error on that one before restarting, but when it came back on
it said 0x8007025D - 0x2000C
The installation failed in the SAFE_OS phase with an error during
APPLY-IMAGE operation.

Today, the damn thing ran fine all day until about 5 PM. The above
erroe suggested faulty install media, so I got a new usb card and made a
new set of install files and tried again. This time it rebooted several
times and got to 82% and screen said the stop code was MEMORY
MANAGEMENT. Upon restarting the scren said 0xC1900101 - 0x4000D
The installation failed in the SECOND BOOT phase with an error during
MIGRATE DATA operation.

It has been running since and I'm typing this on it now.

My questions:
Am I missing something I should do?
Do you think this is a software issue, or hardware?

I would like to fix it if it is a screwed up install or corrupted files,
but I don't really want to do something like try new memory on this old box.

Any help appreciated!


--
Science doesn't support Darwin. Scientists do.


--- PyGate Linux v1.5
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 19/11/2025 12:26 pm, sticks wrote:

The box in question is a rather older machine, but has always been
steady and actually the one I use the most because it is out in my
workshop area. That said, it was originally win 7 that I installed win
10 on several years ago. It runs all day long usually, and I put it to
sleep at night. Not as fast as my newer one in my office, but it works nicely with latest Firefox, BBird, Office 2019 and most other programs I need.

So, a couple days ago, I would come to the computer and notice it had rebooted. The arrangement on my Core Temp taskbar icons were not as I
had left them, and the BBird was closed. So after a couple of these, I started investigating.

Windows Update might automatically reboot Win 10 to update itself. You
might try disabling automatic update.

And... I don't trust nor use the Sleep function.

--
@~@ Simplicity is Beauty! Remain silent! Drink, Blink, Stretch!
/ v \ May the Force and farces be with you! Live long and prosper!!
/( _ )\ https://sites.google.com/site/changmw/
^ ^ https://github.com/changmw/changmw

--- PyGate Linux v1.5
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On Tue, 11/18/2025 11:26 PM, sticks wrote:

The box in question is a rather older machine, but has always been steady and
actually the one I use the most because it is out in my workshop area.ÿ That said,
it was originally win 7 that I installed win 10 on several years ago.ÿ It runs
all day long usually, and I put it to sleep at night.ÿ Not as fast as my newer
one in my office, but it works nicely with latest Firefox, BBird, Office 2019
and most other programs I need.

So, a couple days ago, I would come to the computer and notice it had rebooted.
The arrangement on my Core Temp taskbar icons were not as I had left them, and
the BBird was closed.ÿ So after a couple of these, I started investigating.

Event Viewer showed it booted because of a bugcheck error.ÿ This error was the
same for several reboots.ÿ It didn't reboot while I was using it, and otherwise
worked just fine.

So I did the DISM /restore which couldn't find a source.ÿ It did say the files
were repairable.ÿ So I got an win 10 iso and used it for the source and it did
say it repaired everything.

I then did an SFC/ scannow and it reported everything OK.
Next I did a chkdsk /f and it reported nothing to fix.

Used it for awhile and all seemed ok.ÿ Came back later and it had rebooted again.

Checked on video driver and already had the best one.
Virus check and nothing there

Rebooted again, so I decided to try a system restore point.ÿ First time, it rebooted after a short time doing the restore.

Did a memory check using the windows memory test.ÿ I know it's not the best, but it said all was good.

Tried a second time to restore and after over 2 hours I finally shut it down and restarted.ÿ Hated doing that, but it seemed to be stuck.

Decided to do a restore, saving my files, and it got to the first reboot at about 12% and failed, resetting back to before the setup.ÿ I didn't catch the
error on that one before restarting, but when it came back on it said 0x8007025D - 0x2000C The installation failed in the SAFE_OS phase with an error during APPLY-IMAGE operation.

Today, the damn thing ran fine all day until about 5 PM.ÿ The above error suggested faulty install media, so I got a new usb card and made a new set of install files and tried again.ÿ This time it rebooted several times and got to 82% and screen said the stop code was MEMORY MANAGEMENT.ÿ Upon restarting the screen said 0xC1900101 - 0x4000D The installation failed in the SECOND BOOT
phase with an error during MIGRATE DATA operation.

It has been running since and I'm typing this on it now.

My questions:
Am I missing something I should do?
Do you think this is a software issue, or hardware?

I would like to fix it if it is a screwed up install or corrupted files, but I don't really want to do something like try new memory on this old box.

Any help appreciated!

The first observation, is this is a workshop area PC.

Do you generate a lot of sawdust in the area ? Do you smoke ?
I'm looking for evidence the cooling system is not in good shape,
or the HDD breathe-able hepafilter has ingested a solvent
in the air, or smoke has gummed up the equipment.

Your post makes no reference to the storage device, how many
years it has been spinning all day.

A general idea of make/model of machine would help ascertain
what era of memory it uses, and whether we should be concerned
about memory. DDR2 seemed to be error prone (I only had one
machine which was rock solid on DDR2, and it ran at DDR2-533
which tells you "speed makes a difference" there). DDR3 generally
seems to be pretty good. I have a couple machines on DDR3 and
don't see the failure syndrome you are seeing.

*******

If we go on symptoms, it suggests the "ability to do general computing"
is somehow degraded. It's wobbling when in a low power state.

On relatively old hardware, the CPU was only tested in two PStates,
a high and a low state, the CPU is margined (the margin used to be
500MHz as a lifetime allocation for degradation of unknown sorts).

Modern CPUs have to be tested at more voltage and frequency
conditions, and there are no signs of incompetence when it
comes to SCHMOO plots or the like. Perfectly agile now,
also so agile they change power draw in microseconds.

*******

"it booted because of a bugcheck error"

What was the bugcheck error ?

Is each bugcheck error exactly the same number ?

Or are the bugchecks unrelated, and some of the bugchecks make
no sense at all ?

Bad memory could cause random bugcheck values.

A "hardware issue" could cause the same consistent bugcheck to
show up each time.

But you might also see application errors, if it was RAM.

Where in memory, the RAM problem is, makes a big difference.
If the RAM is bad in the area used to buffer storage, the
symptoms can be quite variable. Bad RAM in application area,
the symptoms might only be applications are tipping over
and crashing out.

And it doesn't matter who makes the RAM, for some kinds of
failures. While most of the time, name brand RAM is good,
I've also had a Crucial stick have one chip completely die on it.

If you have four DIMMs, you can experiment with your DIMM
population. Let us make a diagram of a channel

CPU ===> Swap DIMMs CPU
| on a channel |
High memory DIMM Low memory DIMM
| |
Low memory DIMM High memory DIMM

when you do a memory test of those two configurations, the
test can be more thorough, as any "hard to reach areas"
can be tested by the second test configuration.

With modern RAM, quite frequently the DIMM has such capacity,
you can run with one DIMM, two DIMMs, or all four DIMMs and
the OS can still boot. You can attempt to fault isolate that
way. I had a problem once, where the only problem showed
with all four plugged in, and I couldn't get it to fail
with any other combo. I ended up replacing all four DIMMs
(because they sucked anyway and the RAM was running hotter
than I expected RAM to run). The RAM in that case was not
overvolted.

The Windows Memory Test is actually pretty good, for a mysterious
test method. It could spot a problem, before memtest caught it,
and it took a lot longer memtest to determine there was a problem.
The Test may not visit all memory locations, but for what it is,
it is sufficiently good for our purposes of knowing when
to replace RAM.

But we really need to get back to the symptom set first,
and see whether the bugcheck is pointing at a specific thing,
or the bugchecks are all over the place. Which could be
RAM or CPU.

You should review the SMART statistics on your hard drive,
in case it is on the way out. Age means nothing to a hard drive,
as you can have troubles with them before they hit 10,000 hours,
while other models can do 50,000 hours like it is nothing.
I use the free version of HDTune for that, but there are
other programs which review SMART health.

https://www.smartmontools.org/wiki/Download

# command line, linux-like namespace ?

smartctl -a /dev/sda

# That would check Disk 0 on Windows, /dev/sdb is Disk 1

Paul

--- PyGate Linux v1.5
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 11/19/2025 8:08 AM, Paul wrote:

On Tue, 11/18/2025 11:26 PM, sticks wrote:

The box in question is a rather older machine, but has always been steady and
actually the one I use the most because it is out in my workshop area.ÿ That said,
it was originally win 7 that I installed win 10 on several years ago.ÿ It runs
all day long usually, and I put it to sleep at night.ÿ Not as fast as my newer
one in my office, but it works nicely with latest Firefox, BBird, Office 2019
and most other programs I need.

So, a couple days ago, I would come to the computer and notice it had rebooted.
The arrangement on my Core Temp taskbar icons were not as I had left them, and
the BBird was closed.ÿ So after a couple of these, I started investigating. >>
Event Viewer showed it booted because of a bugcheck error.ÿ This error was the
same for several reboots.ÿ It didn't reboot while I was using it, and otherwise
worked just fine.

So I did the DISM /restore which couldn't find a source.ÿ It did say the files
were repairable.ÿ So I got an win 10 iso and used it for the source and it did
say it repaired everything.

I then did an SFC/ scannow and it reported everything OK.
Next I did a chkdsk /f and it reported nothing to fix.

Used it for awhile and all seemed ok.ÿ Came back later and it had rebooted again.

Checked on video driver and already had the best one.
Virus check and nothing there

Rebooted again, so I decided to try a system restore point.ÿ First time, it >> rebooted after a short time doing the restore.

Did a memory check using the windows memory test.ÿ I know it's not the best, >> but it said all was good.

Tried a second time to restore and after over 2 hours I finally shut it down >> and restarted.ÿ Hated doing that, but it seemed to be stuck.

Decided to do a restore, saving my files, and it got to the first reboot at >> about 12% and failed, resetting back to before the setup.ÿ I didn't catch the
error on that one before restarting, but when it came back on it said
0x8007025D - 0x2000C The installation failed in the SAFE_OS phase with an
error during APPLY-IMAGE operation.

Today, the damn thing ran fine all day until about 5 PM.ÿ The above error
suggested faulty install media, so I got a new usb card and made a new set of
install files and tried again.ÿ This time it rebooted several times and got >> to 82% and screen said the stop code was MEMORY MANAGEMENT.ÿ Upon restarting >> the screen said 0xC1900101 - 0x4000D The installation failed in the SECOND BOOT
phase with an error during MIGRATE DATA operation.

It has been running since and I'm typing this on it now.

My questions:
Am I missing something I should do?
Do you think this is a software issue, or hardware?

I would like to fix it if it is a screwed up install or corrupted files, but >> I don't really want to do something like try new memory on this old box.

Any help appreciated!

The first observation, is this is a workshop area PC.

Do you generate a lot of sawdust in the area ? Do you smoke ?
I'm looking for evidence the cooling system is not in good shape,
or the HDD breathe-able hepafilter has ingested a solvent
in the air, or smoke has gummed up the equipment.

I call it a workshop, but it is actually a small portion of my garage. Originally I wanted a computer in there so I could follow along on
repair videos for things I was working on, like my motorcycle.
I use Core Temp to keep an eye on temperatures as though I heat it, I do
not cool it in the summer. It does get hot for probably about 10 days
each year where the fan really kicks up like when watching a video. But
now, it runs very cool. I did reapply the thermo paste about two years
ago, and if you remember I doubled the ram from 4-8 Gig with your help,
even though the book on it said four was the limit.

Your post makes no reference to the storage device, how many
years it has been spinning all day.

I Like to keep it running because it filters all the incoming crap in my
email so I don't have to deal with it during the day when I look at the
phone. It has been only a few years I keep it running, and I do sleep
it when it gets hot. But I'll explain in a minute why I'm delaying
looking into this for now.

A general idea of make/model of machine would help ascertain
what era of memory it uses, and whether we should be concerned
about memory. DDR2 seemed to be error prone (I only had one
machine which was rock solid on DDR2, and it ran at DDR2-533
which tells you "speed makes a difference" there). DDR3 generally
seems to be pretty good. I have a couple machines on DDR3 and
don't see the failure syndrome you are seeing.

It's a Compaq Presario BM412AA-ABA CQ5600Y
8GB of DDR2. Pegatron Narra5 motherboard.
AMD Phenom II X2 B59 (CAllisto) processor
It's pretty "adult" now.

---snip---

That said, here's where I'm at now. I read something about a failure on
an upgrade perhaps being helped by using the windows upgrade tool. So I
did run it and it found three of what seemed relatively minor fixes.
This is after it had been running three hours without any problems today.

I decided to try and reinstall one more time. (Side Note: Firefox was minimized just now and I was informed it stopped working and closed
Hmmm) Anyways, it finally completed a fresh install taking an hour and
a half to complete. I made a couple adjustments and then immediately
imaged the drive. It has been running now for 8 hours and right up
until Firefox just closed, I have had no problems.

I feel I am at least now in a spot where I can decide if it is hardware
or software related. When I step away from now on, I will close my open programs and if it boots on a fresh install, I can figure it is probably hardware related. If after a couple days I get no problems, I'll leave
the mail open and then the browser, both usually minimized, and go from
there.

FWIW, I'm not against getting a new system, but with the summer heat
this is not the best environment for a computer and if I can keep this
one running without wasting a lot of time, I'd prefer to do that.

Thanks for your help and thoughts. I'll keep them handy in case

--
Science doesn't support Darwin. Scientists do.

--- PyGate Linux v1.5
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On Wed, 11/19/2025 5:04 PM, sticks wrote:

On 11/19/2025 8:08 AM, Paul wrote:

On Tue, 11/18/2025 11:26 PM, sticks wrote:

The box in question is a rather older machine, but has always been steady and
actually the one I use the most because it is out in my workshop area.ÿ That said,
it was originally win 7 that I installed win 10 on several years ago.ÿ It runs
all day long usually, and I put it to sleep at night.ÿ Not as fast as my newer
one in my office, but it works nicely with latest Firefox, BBird, Office 2019
and most other programs I need.

So, a couple days ago, I would come to the computer and notice it had rebooted.
The arrangement on my Core Temp taskbar icons were not as I had left them, and
the BBird was closed.ÿ So after a couple of these, I started investigating. >>>
Event Viewer showed it booted because of a bugcheck error.ÿ This error was the
same for several reboots.ÿ It didn't reboot while I was using it, and otherwise
worked just fine.

So I did the DISM /restore which couldn't find a source.ÿ It did say the files
were repairable.ÿ So I got an win 10 iso and used it for the source and it did
say it repaired everything.

I then did an SFC/ scannow and it reported everything OK.
Next I did a chkdsk /f and it reported nothing to fix.

Used it for awhile and all seemed ok.ÿ Came back later and it had rebooted again.

Checked on video driver and already had the best one.
Virus check and nothing there

Rebooted again, so I decided to try a system restore point.ÿ First time, it >>> rebooted after a short time doing the restore.

Did a memory check using the windows memory test.ÿ I know it's not the best,
but it said all was good.

Tried a second time to restore and after over 2 hours I finally shut it down
and restarted.ÿ Hated doing that, but it seemed to be stuck.

Decided to do a restore, saving my files, and it got to the first reboot at >>> about 12% and failed, resetting back to before the setup.ÿ I didn't catch the
error on that one before restarting, but when it came back on it said
0x8007025D - 0x2000C The installation failed in the SAFE_OS phase with an >>> error during APPLY-IMAGE operation.

Today, the damn thing ran fine all day until about 5 PM.ÿ The above error >>> suggested faulty install media, so I got a new usb card and made a new set of
install files and tried again.ÿ This time it rebooted several times and got >>> to 82% and screen said the stop code was MEMORY MANAGEMENT.ÿ Upon restarting
the screen said 0xC1900101 - 0x4000D The installation failed in the SECOND BOOT
phase with an error during MIGRATE DATA operation.

It has been running since and I'm typing this on it now.

My questions:
Am I missing something I should do?
Do you think this is a software issue, or hardware?

I would like to fix it if it is a screwed up install or corrupted files, but
I don't really want to do something like try new memory on this old box. >>>
Any help appreciated!

The first observation, is this is a workshop area PC.

Do you generate a lot of sawdust in the area ? Do you smoke ?
I'm looking for evidence the cooling system is not in good shape,
or the HDD breathe-able hepafilter has ingested a solvent
in the air, or smoke has gummed up the equipment.

I call it a workshop, but it is actually a small portion of my garage. Originally I wanted a computer in there so I could follow along on repair videos for things I was working on, like my motorcycle.
I use Core Temp to keep an eye on temperatures as though I heat it, I do not cool it in the summer.ÿ It does get hot for probably about 10 days each year where the fan really kicks up like when watching a video.ÿ But now, it runs very cool.ÿ I did reapply the thermo paste about two years ago, and if you remember I doubled the ram from 4-8 Gig with your help, even though the book on it said four was the limit.

Your post makes no reference to the storage device, how many
years it has been spinning all day.ÿ

I Like to keep it running because it filters all the incoming crap in my email so I don't have to deal with it during the day when I look at the phone.ÿ It has been only a few years I keep it running, and I do sleep it when it gets hot.ÿ But I'll explain in a minute why I'm delaying looking into this for now.

A general idea of make/model of machine would help ascertain
what era of memory it uses, and whether we should be concerned
about memory. DDR2 seemed to be error prone (I only had one
machine which was rock solid on DDR2, and it ran at DDR2-533
which tells you "speed makes a difference" there). DDR3 generally
seems to be pretty good. I have a couple machines on DDR3 and
don't see the failure syndrome you are seeing.

It's a Compaq Presario BM412AA-ABA CQ5600Y
8GB of DDR2.ÿ Pegatron Narra5 motherboard.
AMD Phenom II X2 B59 (CAllisto) processor
It's pretty "adult" now.

---snip---

That said, here's where I'm at now.ÿ I read something about a failure on an upgrade perhaps being helped by using the windows upgrade tool.ÿ So I did run it and it found three of what seemed relatively minor fixes. This is after it had been running three hours without any problems today.

I decided to try and reinstall one more time.ÿ (Side Note:ÿ Firefox was minimized just now and I was informed it stopped working and closed Hmmm)ÿ Anyways, it finally completed a fresh install taking an hour and a half to complete.ÿ I made a couple adjustments and then immediately imaged the drive.ÿ It has been running now for 8 hours and right up until Firefox just closed, I have had no problems.

I feel I am at least now in a spot where I can decide if it is hardware or software related.ÿ When I step away from now on, I will close my open programs and if it boots on a fresh install, I can figure it is probably hardware related.ÿ If after a couple days I get no problems, I'll leave the mail open and then the browser, both usually minimized, and go from there.

FWIW, I'm not against getting a new system, but with the summer heat this is not the best environment for a computer and if I can keep this one running without wasting a lot of time, I'd prefer to do that.

Thanks for your help and thoughts.ÿ I'll keep them handy in case

Picture of the motherboard.

https://theretroweb.com/motherboard/manual/hp-narra5-english-6519ee7637baf969984176.pdf

Dual channel, one DIMM per channel, DDR2, 2GB DIMMs. The chipset chip
may be converting HT (Hypertransport) to PCIe. The processor has the
memory controller, which considering the chipset maker is NVidia, it's
a good thing the processor drives the memory :-) NVidia had some troubles driving memory with a couple of its separate chipsets.

A bit on the CPU.

https://www.cpu-world.com/CPUs/K10/AMD-Phenom%20II%20X2%20B59%20-%20HDXB59WFK2DGM.html

Released somewhere around 2011 or so, so 14 years old. Not really that old.
80 watt processor. 2 cores, 3.4GHz .

Memory type is motherboard-specific (motherboard design decides it).

Supported memory: DDR2-1066, DDR3-1333

Based on my experience here, you'd want the DDR3 option,
but the DDR2 boards would have been cheaper to populate with RAM
(which is why when I had that choice, I chose the
wrong one, which was the DDR2). In exchange for my cheapness
at the time, the system ended up using three sets of four
sticks, and when the motherboard croaked, the third set of
RAM was still good. If I'd used DDR3, I likely would
have only needed the one set of RAM (no mortality on
DDR3 here, at all). Since you only have one DIMM per channel,
that improves the electrical properties of the situation.

The board has quite a few electrolytics on it. I would
visually inspect the pressure relief on the top of them,
for leakage.

Your board also uses the vertical SATA connectors. If those
are the first generation connectors (I have a board with
two of those), you can rip the SATA connector right out of the
motherboard by tugging on it. Later SATA are swaged into the
board to make them more mechanically secure. The first SATA,
the objective was to use "minimum amounts of plastic", which
was a mistake. On mine, I could tell just by looking at them,
that there would be trouble.

Paul

--- PyGate Linux v1.5
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 11/19/2025 6:57 PM, Paul wrote:

On Wed, 11/19/2025 5:04 PM, sticks wrote:

On 11/19/2025 8:08 AM, Paul wrote:

On Tue, 11/18/2025 11:26 PM, sticks wrote:

The box in question is a rather older machine, but has always been steady and
actually the one I use the most because it is out in my workshop area.ÿ That said,
it was originally win 7 that I installed win 10 on several years ago.ÿ It runs
all day long usually, and I put it to sleep at night.ÿ Not as fast as my newer
one in my office, but it works nicely with latest Firefox, BBird, Office 2019
and most other programs I need.

So, a couple days ago, I would come to the computer and notice it had rebooted.
The arrangement on my Core Temp taskbar icons were not as I had left them, and
the BBird was closed.ÿ So after a couple of these, I started investigating.

Event Viewer showed it booted because of a bugcheck error.ÿ This error was the
same for several reboots.ÿ It didn't reboot while I was using it, and otherwise
worked just fine.

So I did the DISM /restore which couldn't find a source.ÿ It did say the files
were repairable.ÿ So I got an win 10 iso and used it for the source and it did
say it repaired everything.

I then did an SFC/ scannow and it reported everything OK.
Next I did a chkdsk /f and it reported nothing to fix.

Used it for awhile and all seemed ok.ÿ Came back later and it had rebooted again.

Checked on video driver and already had the best one.
Virus check and nothing there

Rebooted again, so I decided to try a system restore point.ÿ First time, it
rebooted after a short time doing the restore.

Did a memory check using the windows memory test.ÿ I know it's not the best,
but it said all was good.

Tried a second time to restore and after over 2 hours I finally shut it down
and restarted.ÿ Hated doing that, but it seemed to be stuck.

Decided to do a restore, saving my files, and it got to the first reboot at
about 12% and failed, resetting back to before the setup.ÿ I didn't catch the
error on that one before restarting, but when it came back on it said
0x8007025D - 0x2000C The installation failed in the SAFE_OS phase with an >>>> error during APPLY-IMAGE operation.

Today, the damn thing ran fine all day until about 5 PM.ÿ The above error >>>> suggested faulty install media, so I got a new usb card and made a new set of
install files and tried again.ÿ This time it rebooted several times and got
to 82% and screen said the stop code was MEMORY MANAGEMENT.ÿ Upon restarting
the screen said 0xC1900101 - 0x4000D The installation failed in the SECOND BOOT
phase with an error during MIGRATE DATA operation.

It has been running since and I'm typing this on it now.

My questions:
Am I missing something I should do?
Do you think this is a software issue, or hardware?

I would like to fix it if it is a screwed up install or corrupted files, but
I don't really want to do something like try new memory on this old box. >>>>
Any help appreciated!

The first observation, is this is a workshop area PC.

Do you generate a lot of sawdust in the area ? Do you smoke ?
I'm looking for evidence the cooling system is not in good shape,
or the HDD breathe-able hepafilter has ingested a solvent
in the air, or smoke has gummed up the equipment.

I call it a workshop, but it is actually a small portion of my garage. Originally I wanted a computer in there so I could follow along on repair videos for things I was working on, like my motorcycle.
I use Core Temp to keep an eye on temperatures as though I heat it, I do not cool it in the summer.ÿ It does get hot for probably about 10 days each year where the fan really kicks up like when watching a video.ÿ But now, it runs very cool.ÿ I did reapply the thermo paste about two years ago, and if you remember I doubled the ram from 4-8 Gig with your help, even though the book on it said four was the limit.

Your post makes no reference to the storage device, how many
years it has been spinning all day.

I Like to keep it running because it filters all the incoming crap in my email so I don't have to deal with it during the day when I look at the phone.ÿ It has been only a few years I keep it running, and I do sleep it when it gets hot.ÿ But I'll explain in a minute why I'm delaying looking into this for now.

A general idea of make/model of machine would help ascertain
what era of memory it uses, and whether we should be concerned
about memory. DDR2 seemed to be error prone (I only had one
machine which was rock solid on DDR2, and it ran at DDR2-533
which tells you "speed makes a difference" there). DDR3 generally
seems to be pretty good. I have a couple machines on DDR3 and
don't see the failure syndrome you are seeing.

It's a Compaq Presario BM412AA-ABA CQ5600Y
8GB of DDR2.ÿ Pegatron Narra5 motherboard.
AMD Phenom II X2 B59 (CAllisto) processor
It's pretty "adult" now.

---snip---

That said, here's where I'm at now.ÿ I read something about a failure on an upgrade perhaps being helped by using the windows upgrade tool.ÿ So I did run it and it found three of what seemed relatively minor fixes. This is after it had been running three hours without any problems today.

I decided to try and reinstall one more time.ÿ (Side Note:ÿ Firefox was minimized just now and I was informed it stopped working and closed Hmmm)ÿ Anyways, it finally completed a fresh install taking an hour and a half to complete.ÿ I made a couple adjustments and then immediately imaged the drive.ÿ It has been running now for 8 hours and right up until Firefox just closed, I have had no problems.

I feel I am at least now in a spot where I can decide if it is hardware or software related.ÿ When I step away from now on, I will close my open programs and if it boots on a fresh install, I can figure it is probably hardware related.ÿ If after a couple days I get no problems, I'll leave the mail open and then the browser, both usually minimized, and go from there.

FWIW, I'm not against getting a new system, but with the summer heat this is not the best environment for a computer and if I can keep this one running without wasting a lot of time, I'd prefer to do that.

Thanks for your help and thoughts.ÿ I'll keep them handy in case

Picture of the motherboard.

https://theretroweb.com/motherboard/manual/hp-narra5-english-6519ee7637baf969984176.pdf

Dual channel, one DIMM per channel, DDR2, 2GB DIMMs. The chipset chip
may be converting HT (Hypertransport) to PCIe. The processor has the
memory controller, which considering the chipset maker is NVidia, it's
a good thing the processor drives the memory :-) NVidia had some troubles driving memory with a couple of its separate chipsets.

A bit on the CPU.

https://www.cpu-world.com/CPUs/K10/AMD-Phenom%20II%20X2%20B59%20-%20HDXB59WFK2DGM.html

Released somewhere around 2011 or so, so 14 years old. Not really that old. 80 watt processor. 2 cores, 3.4GHz .

Memory type is motherboard-specific (motherboard design decides it).

Supported memory: DDR2-1066, DDR3-1333

Based on my experience here, you'd want the DDR3 option,
but the DDR2 boards would have been cheaper to populate with RAM
(which is why when I had that choice, I chose the
wrong one, which was the DDR2). In exchange for my cheapness
at the time, the system ended up using three sets of four
sticks, and when the motherboard croaked, the third set of
RAM was still good. If I'd used DDR3, I likely would
have only needed the one set of RAM (no mortality on
DDR3 here, at all). Since you only have one DIMM per channel,
that improves the electrical properties of the situation.

The board has quite a few electrolytics on it. I would
visually inspect the pressure relief on the top of them,
for leakage.

Your board also uses the vertical SATA connectors. If those
are the first generation connectors (I have a board with
two of those), you can rip the SATA connector right out of the
motherboard by tugging on it. Later SATA are swaged into the
board to make them more mechanically secure. The first SATA,
the objective was to use "minimum amounts of plastic", which
was a mistake. On mine, I could tell just by looking at them,
that there would be trouble.

I bought the thing several years ago, so long I can't remember actually,
but I am pretty sure I only gave $50 for it. I used to go wireless, but
it's pretty far from the router so I wired out here for ethernet and
that helped with function a lot. For what my original intentions were
for having something out in the garage, this damn thing has been pretty
good. This is really the first time I've had an issue, and it appears something got corrupted, but just might have been fixed with the
successful reinstall. All day now with no reboots. Fingers crossed.


--
Science doesn't support Darwin. Scientists do.

--- PyGate Linux v1.5.1
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 11/19/2025 8:52 PM, sticks wrote:

On 11/19/2025 6:57 PM, Paul wrote:

On Wed, 11/19/2025 5:04 PM, sticks wrote:

---snip---

Your post makes no reference to the storage device, how many
years it has been spinning all day.

https://i.postimg.cc/4dyW4QHp/Smart.jpg

I bought the thing several years ago, so long I can't remember actually,
but I am pretty sure I only gave $50 for it.ÿ I used to go wireless, but it's pretty far from the router so I wired out here for ethernet and
that helped with function a lot.ÿ For what my original intentions were
for having something out in the garage, this damn thing has been pretty good.ÿ This is really the first time I've had an issue, and it appears something got corrupted, but just might have been fixed with the
successful reinstall.ÿ All day now with no reboots.ÿ Fingers crossed.

It started rebooting again, so I reran the chkdsk, sfc, and dism stuff,
but it still had problems. Below were some of the events. They were
not all the same, and I remembered your advice on checking out the disk
so I installed HD tune to check it out, and got the results in the link
above. It does look like the drive might be on it's last legs.

So before I just go and buy a new box, I'm gonna try another drive I
have laying around. It's a 640 Gig laptop drive, but I'm not buying a
new desktop drive for this old machine. I have a brand new 1 TB Samsung
990, but I can't use it since it doesn't have the connectors on the
board for it. If the spinner works without errors, I'll know that was
the problem. Fun fun fun.

----------------------
1. The computer has rebooted from a bugcheck. The bugcheck was:
0x00000139 (0x0000000000000003, 0xffffb783ade056c0, 0xffffb783ade05618, 0x0000000000000000).

2. The computer has rebooted from a bugcheck. The bugcheck was:
0x0000003b (0x00000000c0000005, 0xfffff80627a83518, 0xfffff8062b0ad920, 0x0000000000000000).
The computer has rebooted from a bugcheck. The bugcheck was: 0x0000003b (0x00000000c0000005, 0xfffff8003fb08db2, 0xffff8e88cd64ba30, 0x0000000000000000).

3. The computer has rebooted from a bugcheck. The bugcheck was:
0x0000003b (0x00000000c0000005, 0xfffff8003fb08db2, 0xffff8e88cd64ba30, 0x0000000000000000).

4. The computer has rebooted from a bugcheck. The bugcheck was:
0x0000001e (0xffffffffc0000005, 0xfffff800528d23c5, 0x0000000000000001, 0x0000000000000103).

--- PyGate Linux v1.5.1
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On Mon, 11/24/2025 5:54 PM, sticks wrote:

On 11/19/2025 8:52 PM, sticks wrote:

On 11/19/2025 6:57 PM, Paul wrote:

On Wed, 11/19/2025 5:04 PM, sticks wrote:

---snip---

Your post makes no reference to the storage device, how many
years it has been spinning all day.

https://i.postimg.cc/4dyW4QHp/Smart.jpg

I bought the thing several years ago, so long I can't remember actually, but I am pretty sure I only gave $50 for it.ÿ I used to go wireless, but it's pretty far from the router so I wired out here for ethernet and that helped with function a lot.ÿ For what my original intentions were for having something out in the garage, this damn thing has been pretty good.ÿ This is really the first time I've had an issue, and it appears something got corrupted, but just might have been fixed with the successful reinstall.ÿ All day now with no reboots.ÿ Fingers crossed.

It started rebooting again, so I reran the chkdsk, sfc, and dism stuff, but it still had problems.ÿ Below were some of the events.ÿ They were not all the same, and I remembered your advice on checking out the disk so I installed HD tune to check it out, and got the results in the link above.ÿ It does look like the drive might be on it's last legs.

So before I just go and buy a new box, I'm gonna try another drive I have laying around.ÿ It's a 640 Gig laptop drive, but I'm not buying a new desktop drive for this old machine.ÿ I have a brand new 1 TB Samsung 990, but I can't use it since it doesn't have the connectors on the board for it.ÿ If the spinner works without errors, I'll know that was the problem.ÿ Fun fun fun.

----------------------
1.ÿ The computer has rebooted from a bugcheck.ÿ The bugcheck was: 0x00000139 (0x0000000000000003, 0xffffb783ade056c0, 0xffffb783ade05618, 0x0000000000000000).

2.ÿ The computer has rebooted from a bugcheck.ÿ The bugcheck was: 0x0000003b (0x00000000c0000005, 0xfffff80627a83518, 0xfffff8062b0ad920, 0x0000000000000000).
The computer has rebooted from a bugcheck.ÿ The bugcheck was: 0x0000003b (0x00000000c0000005, 0xfffff8003fb08db2, 0xffff8e88cd64ba30, 0x0000000000000000).

3.ÿ The computer has rebooted from a bugcheck.ÿ The bugcheck was: 0x0000003b (0x00000000c0000005, 0xfffff8003fb08db2, 0xffff8e88cd64ba30, 0x0000000000000000).

4.ÿ The computer has rebooted from a bugcheck.ÿ The bugcheck was: 0x0000001e (0xffffffffc0000005, 0xfffff800528d23c5, 0x0000000000000001, 0x0000000000000103).

Your Reallocated Sector Count seems a bit strange.
I think it is trying to report "2", at a guess.

Your bugchecks seem to be in Ring0, which could be memory used
for kernel and drivers. That does not have a strong disk drive
component to it, directly. And while it could have an indirect
effect, you would think there would be more splatter in the
form of crashing or erroneous application behavior. Your error
behavior seems relatively focused.

KERNEL_SECURITY_CHECK_FAILURE bug check has a value of 0x00000139

SYSTEM_SERVICE_EXCEPTION bug check has a value of 0x0000003B (possibly driver related)

KMODE_EXCEPTION_NOT_HANDLED bug check has a value of 0x0000001E (status_access_violation as well)

I had a weird one in WinXP days, that "seemed" to be ram buffer used for
disk operation (kernel level). When the RAM was changed out, the problem was gone for
a while, but I don't really think it was gone. Now that the machine
is dead, no other machine in the room has had such symptoms.
Not every one of these problems, "behaves cleanly". They
do all sorts of weird stuff.

*******

By all means, try another drive. Just don't clone it on
that machine, OK ? Take the two disks to a machine that is
working, and do your cloning there. You may need to run a
CHKDSK using the machine that works well, before setting
up to do a clone. If there is a CRC error during cloning, most
cloning programs stop if that happens. The attempt to clone,
might even be a kind of test.

Paul

--- PyGate Linux v1.5.1
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 11/24/2025 6:36 PM, Paul wrote:

By all means, try another drive. Just don't clone it on
that machine, OK ? Take the two disks to a machine that is
working, and do your cloning there. You may need to run a
CHKDSK using the machine that works well, before setting
up to do a clone. If there is a CRC error during cloning, most
cloning programs stop if that happens. The attempt to clone,
might even be a kind of test.

When I did finally get windows reinstalled on it, I did immediately take
an image of it. My first choice of what to do is to go to my desktop in
the house that has a spinner I can temporarily remove. I'll hook up the laptop drive and just restore that image to it and then see if the
garage machine will boot on it.

If that one seems bad, I do have an image of when I first got it all set
up again about a year ago. I could try that one next, but I'll try the
one from a few days ago with a fresh windows 10 install first.

Thank you

--- PyGate Linux v1.5.1
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 11/24/2025 6:52 PM, sticks wrote:

On 11/24/2025 6:36 PM, Paul wrote:

By all means, try another drive. Just don't clone it on
that machine, OK ? Take the two disks to a machine that is
working, and do your cloning there. You may need to run a
CHKDSK using the machine that works well, before setting
up to do a clone. If there is a CRC error during cloning, most
cloning programs stop if that happens. The attempt to clone,
might even be a kind of test.

When I did finally get windows reinstalled on it, I did immediately take
an image of it.ÿ My first choice of what to do is to go to my desktop in
the house that has a spinner I can temporarily remove.ÿ I'll hook up the laptop drive and just restore that image to it and then see if the
garage machine will boot on it.

If that one seems bad, I do have an image of when I first got it all set
up again about a year ago.ÿ I could try that one next, but I'll try the
one from a few days ago with a fresh windows 10 install first.

Just to give feedback on the whole thing, I'll explain the end result.

I did put the newer drive (900 hours) on it and after struggling for a
few days, decided to just do a fresh install. Even that gave me
problems, and only after disabling video card drivers and a bunch of
other crap did I finally get it to finish a new install. Almost
immediately that was giving me different stopcodes and bugchecks and unexpected reboots. So I decided it most likely was time to throw in
the towel on this 15 year old hardware and get a new box.

I'm using that new system now, and am glad I did it. I wasted over a
whole week screwing around with it. Though I do enjoy fixing things and learning about this stuff, it is getting less appealing to me than it
used to be. I just have far more important things to do than try and
save a few bucks on old hardware.

sticks

--- PyGate Linux v1.5.1
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)