Re: Bad memory

2013-01-18 Thread Frank Murphy
On Thu, 17 Jan 2013 10:01:41 -0800
Chuck Forsberg WA7KGX N2469R c...@omen.com wrote:
 
 Option 2 has already been tried.  The old motherboard was checked
 with memtest86+ before being placed in service running omen.com.
 
Maybe use an out of box solution: 
https://www.youtube.com/watch?v=1IomAerbV7g


-- 
Regards,
Frank

Ask not what Fedora can do for you,
  but what you can do for Fedora
--me, courtesy JFK
-- 
test mailing list
test@lists.fedoraproject.org
To unsubscribe:
https://admin.fedoraproject.org/mailman/listinfo/test

Re: Bad memory

2013-01-18 Thread Justin M. Forbes
On Thu, Jan 17, 2013 at 09:18:40AM -0800, Chuck Forsberg WA7KGX N2469R wrote:
 Recently I discovered the motherboard that had been running omen.com
 had an undetected bad memory chip.   This is a real confidence builder.
 
 Currently, running a definitive memory test requires hours of down time.
 The user mode memory tester has serious limitations.
 
 I propose kernel support for a user mode program similar to memtest86
 to allow such a program to request a hardware block of memory and
 relocate whatever was in that block to another memory area.  This way
 all of the memory can be checked in the background of a running system.
 

There are a lof of reasons that this isn't feasible on a running system.
You just don't have access to all of the memory for a full check.

Justin
-- 
test mailing list
test@lists.fedoraproject.org
To unsubscribe:
https://admin.fedoraproject.org/mailman/listinfo/test

Re: Bad memory

2013-01-18 Thread cornel panceac
2013/1/18 Justin M. Forbes jmfor...@linuxtx.org

 On Thu, Jan 17, 2013 at 09:18:40AM -0800, Chuck Forsberg WA7KGX N2469R
 wrote:
  Recently I discovered the motherboard that had been running omen.com
  had an undetected bad memory chip.   This is a real confidence builder.
 
  Currently, running a definitive memory test requires hours of down time.
  The user mode memory tester has serious limitations.
 
  I propose kernel support for a user mode program similar to memtest86
  to allow such a program to request a hardware block of memory and
  relocate whatever was in that block to another memory area.  This way
  all of the memory can be checked in the background of a running system.
 

 There are a lof of reasons that this isn't feasible on a running system.
 You just don't have access to all of the memory for a full check.

 Justin

 why not? is there something that can not be relocated in RAM?

on the other hand, wasn't there supposed to be a way to exclude a damaged
RAM area by passing some arguments to the kernel?
-- 
test mailing list
test@lists.fedoraproject.org
To unsubscribe:
https://admin.fedoraproject.org/mailman/listinfo/test

Re: Bad memory

2013-01-18 Thread cornel panceac
2013/1/18 Adam Jackson a...@redhat.com

 On Fri, 2013-01-18 at 20:49 +0200, cornel panceac wrote:


  2013/1/18 Justin M. Forbes jmfor...@linuxtx.org

  There are a lof of reasons that this isn't feasible on a
  running system.
  You just don't have access to all of the memory for a full
  check.
 
  Justin
 
  why not? is there something that can not be relocated in RAM?

 The kernel image itself.  DMA buffers.  Memory made inaccessible by PCI
 BARs being in the way.  Probably some other stuff I'm forgetting.  Even
 if you fixed all that you'd still be Heisenberg-uncertain because you'd
 no longer be the only thing exerting cache pressure.

  on the other hand, wasn't there supposed to be a way to exclude a
  damaged RAM area by passing some arguments to the kernel?

 See linux/Documentation/bad_memory.txt

 - ajax



 ok, thank you very much, ajax.
-- 
test mailing list
test@lists.fedoraproject.org
To unsubscribe:
https://admin.fedoraproject.org/mailman/listinfo/test

Re: Bad memory

2013-01-17 Thread Frank Murphy
On Thu, 17 Jan 2013 09:18:40 -0800
Chuck Forsberg WA7KGX N2469R c...@omen.com wrote:

 Recently I discovered the motherboard that had been running omen.com
 had an undetected bad memory chip.   This is a real confidence
 builder.
 
 Currently, running a definitive memory test requires hours of down
 time. The user mode memory tester has serious limitations.
 


Option 2, but a stick of ram.
Less fannying about.


-- 
Regards,
Frank

Byte my kernel
--me .
-- 
test mailing list
test@lists.fedoraproject.org
To unsubscribe:
https://admin.fedoraproject.org/mailman/listinfo/test