LinuxQuestions.org
Help answer threads with 0 replies.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Hardware
User Name
Password
Linux - Hardware This forum is for Hardware issues.
Having trouble installing a piece of hardware? Want to know if that peripheral is compatible with Linux?

Notices


Reply
  Search this Thread
Old 03-29-2011, 01:00 PM   #1
hua
Member
 
Registered: Oct 2006
Location: Slovak Republic
Distribution: Slackware 14.2, current
Posts: 461

Rep: Reputation: 78
It is possible to damage system memory by the OS??


I have a new lenovo thinkserver hardware. I was configuring my newly installed kernel (slackware 13.1 - kernel 2.6.33.4) maybe to weeks now. I was testing the configuration of the kernel parameters.

Suddenly out of the blue my BIOS started to report memory error. I have 6 Fully buffered memories in pairs. Four has no advanced ECC and two supports ECC.

It is possible to damage the memory by some wrong kernel configuration???
I cannot believe that this is possible but I cannot see any other explanation how can a two weeks old hardware get wrong.

 
Click here to see the post LQ members have rated as the most helpful post in this thread.
Old 03-29-2011, 02:20 PM   #2
knudfl
LQ 5k Club
 
Registered: Jan 2008
Location: Copenhagen DK
Distribution: PCLinuxOS2023 Fedora38 + 50+ other Linux OS, for test only.
Posts: 17,511

Rep: Reputation: 3641Reputation: 3641Reputation: 3641Reputation: 3641Reputation: 3641Reputation: 3641Reputation: 3641Reputation: 3641Reputation: 3641Reputation: 3641Reputation: 3641
Errors : Also present when the equipment is "first time tested".

The manufacturers test will remove say 99% failures with 97% certainty.

I guess, you have some warranty. Better go get a new unit.
 
1 members found this post helpful.
Old 03-29-2011, 02:36 PM   #3
kilgoretrout
Senior Member
 
Registered: Oct 2003
Posts: 2,983

Rep: Reputation: 388Reputation: 388Reputation: 388Reputation: 388
Quote:
I have 6 Fully buffered memories in pairs. Four has no advanced ECC and two supports ECC
You should be aware that if you add non-ECC memory to a system with ECC memory, the error checking capabilities of the ECC memory will be disabled. Also, make sure that each ram module you are using is compatible with your motherboard. Mixing different types of ram can cause problems on certain motherboards even when each ram stick by itself is compatible.

To answer your question, I don't see how a kernel configuration could damage your ram. There may be certain kernel configurations which may cause the ram to malfunction but this would not result in permanent damage. I think it is more likely you have a ram compatibility problem. If you have altered your bios ram settings from auto/default or have overclocked, you may want to set your bios back to its default settings.
 
2 members found this post helpful.
Old 03-29-2011, 02:38 PM   #4
tredegar
LQ 5k Club
 
Registered: May 2003
Location: London, UK
Distribution: Fedora38
Posts: 6,147

Rep: Reputation: 435Reputation: 435Reputation: 435Reputation: 435Reputation: 435
Quote:
It is possible to damage the memory by some wrong kernel configuration???
Not as far as I know.

Most (all?) hardware is tested before it leaves the factory.

Then there is "Burn in", when it fails within days or weeks. Manufacturers know this.
Then there is "It just works".
Then there is "Burn out", when it fails after years (sometimes many years - I still have a 20y old HDD running fine, but it is backed up, and not important).

Looks like you have a "burn in" problem, and knudfl's advice is good.
 
2 members found this post helpful.
Old 03-29-2011, 03:10 PM   #5
jefro
Moderator
 
Registered: Mar 2008
Posts: 21,939

Rep: Reputation: 3619Reputation: 3619Reputation: 3619Reputation: 3619Reputation: 3619Reputation: 3619Reputation: 3619Reputation: 3619Reputation: 3619Reputation: 3619Reputation: 3619
I don't know exactly what error you see but..

I have been around computers for a while. I am not an engineer but I have never ever heard of any way to damage memory by any code of any sort. I dare someone to prove it can be done. You'd have to understand the way it connects. No amount of code can change the electric situation and cause it to be damaged. It is only remotely possible that if you ran memtest on a box with no fans for a day or week in a very hot room it might be thermally damaged. Remember they go on a wave solder and that is 360 degrees or so or more so in an oven to melt paste.

They tend to fail based on ESD damage in maybe 90% of cases of pre-mature failure. The other were mixture of design and other factors like heat or mechanical damage. This is from a factory report.

In fact almost no hardware is tested at factories any more. Sony was the fist to introduce the customer being the single test. It used to be that at IBM, the PS/2/s were run in hot and cold chambers for 24 hours. By the time IBM sold to lenovo almost no tests were being performed other than samples and a few bed of nails tests. I doubt lenovo has done any line testing. No company other than very high end enterprise level system do any burn in tests. Only the very highest end systems and supercomputers get tested anymore. Consumer junk it simply too low of a profit margin to do any real testing.

Last edited by jefro; 03-29-2011 at 03:15 PM.
 
Old 03-29-2011, 07:01 PM   #6
jlinkels
LQ Guru
 
Registered: Oct 2003
Location: Bonaire, Leeuwarden
Distribution: Debian /Jessie/Stretch/Sid, Linux Mint DE
Posts: 5,195

Rep: Reputation: 1043Reputation: 1043Reputation: 1043Reputation: 1043Reputation: 1043Reputation: 1043Reputation: 1043Reputation: 1043
No. But if you use a different OS and you have a faulty memory it may show up, while it doesn't with other OS-es or versions.

jlinkels
 
Old 03-30-2011, 07:18 AM   #7
hua
Member
 
Registered: Oct 2006
Location: Slovak Republic
Distribution: Slackware 14.2, current
Posts: 461

Original Poster
Rep: Reputation: 78
Thank you for your answers.
I was expecting hardware problem too but I thought that it is normal only in the desktop-hardware environment. I mean that the manufacturer ships non-tested hardware only for desktops.

Quote:
Originally Posted by tredegar View Post
Then there is "Burn in", when it fails within days or weeks. Manufacturers know this.
Then there is "It just works".
Then there is "Burn out", when it fails after years
Thanks for this, it is very helpful.
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
File recovery after partition table damage - XFS file system gracecourt Linux - General 4 09-28-2017 05:29 PM
Possible file-system damage? hacker supreme General 11 02-03-2008 10:36 AM
Difference between resident memory,shared memory and virtual memory in system monitor mathimca05 Linux - Newbie 1 11-11-2007 04:05 AM
will "skin so soft" damage my system internals? slinky2004 Linux - Hardware 4 09-14-2005 10:02 PM
File system damage ext mandrake 10.0 and VMware gaelen Mandriva 1 04-09-2005 11:09 AM

LinuxQuestions.org > Forums > Linux Forums > Linux - Hardware

All times are GMT -5. The time now is 09:36 PM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Open Source Consulting | Domain Registration