Windows Rot
View previous topic | View next topic >
Post new topic Reply to topic
M.A.I.L. Forum Index -> Chat
   
Author Message

Joined: August 30, 2008
Posts: 3064
Submissions: 20
Location: Burlington, ON, Canada

Reply with quote
Posted on Wed Aug 10, 2011 9:37 pm
Link to Post: Link to Post

Zncon wrote:
epiphany2006 -

That sounds like either a memory or HDD issue, or perhaps both. It's a bit odd that Ubuntu works fine, but it does have a smaller memory footprint, so it may not have hit the bad memory location.


That, or possibly a windows driver having a disagreement with a device.
Said device is either unrecognized by Ubuntu or treated to a rather 'Generic' driver offering limited functionality (often the case with onboard VGA/LAN/wLAN)...
Even if there is a special driver written for it, it may not cause the problems that the Windows Driver does, or access whatever the function is that caused the BSOD's...

Food for thought.

epiphany, did you ever think to look up the error codes the blue screens generated?



Joined: May 07, 2008
Posts: 3615
Submissions: 149
Location: Germany, Herxheim

Reply with quote
Posted on Wed Aug 10, 2011 10:07 pm
Link to Post: Link to Post

Shall I tell you what? During the last days my system showed already some irregularities (system hang-ups, not even producing logged BSODs for analysis) - and today I seem to have found the culprit - a faulty memory module, when some smoke left it. After removal of the memory bank that module belonged to, my system worked again - the Vista install seems to have survived the numerous hangings; not even my RAID array that is in write-back/delayed write mode suffered data loss...

See, that in every electronic component a small amount of smoke is built-in. So if it's not more gas tight, e.g.doe to a cracked case, and the smoke can escape, the component will go defective. Smile

I hope that was the only issue, and Northbridge/MMU didn't suffer by the memory failure - that PC has to work (more or less) reliable for further 15 months, as I calculated this one for a primary life time of fifty months - November 10, 2012 it may retire....

-ZiLi-


Maille Code V2.0 T7.1 R5.6 Ep Fper MAl Ws$ Cpbsw$ G0.3-6.4 I1.0-30.0 N28.25 Ps Dacdejst Xagtw S08 Hip

Human societies are like chain mail.
A single link will be worth nothing.
A chain is of use, but will break at the weakest link.
A weak weave will have the need to replace weak links.
A strong weave will survive even with weak links included.
-'me

Joined: September 15, 2005
Posts: 136
Submissions: 0

Reply with quote
Posted on Wed Aug 10, 2011 10:09 pm
Link to Post: Link to Post

Always have to be careful not to let the magic smoke out.

Joined: February 6, 2011
Posts: 5
Submissions: 0
Location: Arkansas

Reply with quote
Posted on Wed Aug 10, 2011 10:49 pm
Link to Post: Link to Post

Yes, I did look them up but unfortunately it was all gibberish to me. Something about drivers and software but not which ones. There had been nothing new software or hardware wise added other than regular Windows updates. I went thru the system and checked drivers for everything, everything showed working normally, no conflicts. I did the bios memory and disk checks several times, all came back normal. It got to where it couldn't finish startup with out a blue screen. Even starting up in safe mode it would hesitate and restart at times.

Joined: May 07, 2008
Posts: 3615
Submissions: 149
Location: Germany, Herxheim

Reply with quote
Posted on Wed Aug 10, 2011 11:02 pm
Link to Post: Link to Post

epiphany: I share the suspicion zncon already uttered. Do yourself a favor, and make a DOS bootable diskette, usb stick or CD containing the tools MHDD, and Memtest86+. And let the hdd test as well as the memory test run overnight in loop mode, to be halfways sure that you don't have faulty hardware - I had, as reported in my previous post.

And if you have BSODs, it's always good to note the main STOP code, like e.g. 0x0000007F or so, and then google for it (the four bracketed strings are usually less important for a normal user, as they can be analyzed most times only by Pro technicians).

-ZiLi-


Maille Code V2.0 T7.1 R5.6 Ep Fper MAl Ws$ Cpbsw$ G0.3-6.4 I1.0-30.0 N28.25 Ps Dacdejst Xagtw S08 Hip

Human societies are like chain mail.
A single link will be worth nothing.
A chain is of use, but will break at the weakest link.
A weak weave will have the need to replace weak links.
A strong weave will survive even with weak links included.
-'me

Joined: September 15, 2005
Posts: 136
Submissions: 0

Reply with quote
Posted on Wed Aug 10, 2011 11:07 pm
Link to Post: Link to Post

Memtest is one of my favorite tools for this sort of thing. I just find it hard to get other people to use it well.

Joined: February 6, 2011
Posts: 5
Submissions: 0
Location: Arkansas

Reply with quote
Posted on Thu Aug 11, 2011 10:21 pm
Link to Post: Link to Post

Thanks for the tips, I'll try them out when I get around to switching back to Windows.

Joined: September 21, 2009
Posts: 65
Submissions: 1

Reply with quote
Posted on Fri Aug 12, 2011 5:30 pm
Link to Post: Link to Post

In my experience, your issues are consistent with bad memory. It might also be bad blocks under the page file, but that is less likely.

You boot into memtest, so you don't need to wait until you switch back to run it. Some of the LiveCD install CDs have an option to boot into memtest. Note, it doesn't stress the memory so can't find intermittant problems, but yours don't sound intermittant.

Linux also has a disk badblock tester (badblocks?) that's a far sight better than what Windows has, if you want to check the drive. Just don't do any write tests unless you're ready to wipe the disk.


Also, when replacing memory modules, make sure you stay grounded the entire time you're handling the memory modules. Just discharging yourself and then letting go to work on the innards isn't really sufficient. The damage caused usually doesn't reveal itself immediately. A wrist gounding strap is a good idea.

Joined: September 15, 2005
Posts: 136
Submissions: 0

Reply with quote
Posted on Fri Aug 12, 2011 5:49 pm
Link to Post: Link to Post

I think it's time to start a new maille product. Chainmaille grounding bracelets/straps? Small copper JPL would be plenty conductive.

Joined: September 21, 2009
Posts: 65
Submissions: 1

Reply with quote
Posted on Fri Aug 12, 2011 6:06 pm
Link to Post: Link to Post

As I understand it, actual grounding straps have some built in resistance to their grounding path as a safety feature -- that way they don't become a high current path to ground if the user accidently touches some high voltage.

Hmmm.. I sense some multimeter work this evening to find where they've installed that resistance Smile

Joined: September 15, 2005
Posts: 136
Submissions: 0

Reply with quote
Posted on Fri Aug 12, 2011 6:16 pm
Link to Post: Link to Post

Interesting, I'd never heard that before, and Wikipedia agrees with you.

I guess some people just don't like high voltage shocks.

Joined: May 07, 2008
Posts: 3615
Submissions: 149
Location: Germany, Herxheim

Reply with quote
Posted on Fri Aug 12, 2011 6:52 pm
Link to Post: Link to Post

Zncon wrote:
I think it's time to start a new maille product. Chainmaille grounding bracelets/straps? Small copper JPL would be plenty conductive.


I built (and sometimes repaired) several thousands of PCs, with power cord NOT pulled from PSU, but EXPLICITLY switched off, so it was always protective grounded. And as it's inevitable to touch the computer's metallic case when mounting, I was always grounded as well.

But yes, it's clear that it's better to work on a professional conductive (and grounded) surface work bench, connected to it with a conductor wristband, wearing conductive shoes, running around on the workshop's conductive floor. Been there, done that. But at home, or at the customer for a repair, the switched-off PSU with power cord's PG connection works absolutely sufficient.

-ZiLi-


Maille Code V2.0 T7.1 R5.6 Ep Fper MAl Ws$ Cpbsw$ G0.3-6.4 I1.0-30.0 N28.25 Ps Dacdejst Xagtw S08 Hip

Human societies are like chain mail.
A single link will be worth nothing.
A chain is of use, but will break at the weakest link.
A weak weave will have the need to replace weak links.
A strong weave will survive even with weak links included.
-'me

Joined: September 15, 2005
Posts: 136
Submissions: 0

Reply with quote
Posted on Fri Aug 12, 2011 6:57 pm
Link to Post: Link to Post

I have a set up based on a similar concept. I have a standard PC power cord with only the ground plug left intact which I use for repairs I do at home.

Joined: September 21, 2009
Posts: 65
Submissions: 1

Reply with quote
Posted on Fri Aug 12, 2011 7:19 pm
Link to Post: Link to Post

Zili, I know you know this, but others reading this thread may not...

Actually. its not so much the grounding that's important, but that everything is at the same electrical potential. Having everything referenced to ground is the simplest way, and that's what all those professional devices you mentioned accomplish.

Leaving the chassis plugged into the wall doesn't help accomplish this at all. You need to stay at the same electical potential as the chassis, which is easily done by continually touching it while installing the memory module (not that continually touching the chassis is all that easy.. hence the wrist strap recommendation). Might as well just unplug the computer too anyway and avoid any risk of damage from the standby 5V, although the mechanical power switch on the power supply like Zili mentions suffices to eliminate this risk too.

Post new topic Reply to topic
Jump to:  
Page 2 of 2. Goto page Previous  1, 2
All times are GMT. The time now is Tue Sep 24, 2019 9:59 am
M.A.I.L. Forum Index -> Chat
Display posts from previous: