1 |
On Sunday 21 August 2011, Nikos Chantziaras wrote: |
2 |
> On 08/21/2011 02:19 PM, Francesco Talamona wrote: |
3 |
> > I wish yours it's not a RAM |
4 |
> > |
5 |
> > issue, it could be tricky to spot, because memtest is not putting |
6 |
> > any load to the machine, so it's very useful when it reports |
7 |
> > error, but when it doesn't you can't be sure if RAM modules are in |
8 |
> > good health. |
9 |
> |
10 |
> CPU load doesn't affect RAM errors. CPU load affects CPU errors. If |
11 |
> you only get RAM errors during heavy load, the RAM is just fine, but |
12 |
> your CPU has a fault. |
13 |
|
14 |
I see your point: to better explain my statement I point you to |
15 |
http://people.redhat.com/~dledford/memtest.shtml |
16 |
|
17 |
The idea is that a "synthetic" test isn't guaranteed to repeat real life |
18 |
conditions, so its results has to be interpreted rather than taken |
19 |
acritically. |
20 |
|
21 |
Cheers |
22 |
Francesco |
23 |
-- |
24 |
Linux Version 3.0.3-gentoo, Compiled #1 SMP PREEMPT Fri Aug 19 07:16:13 |
25 |
CEST 2011 |
26 |
Two 2.9GHz AMD Athlon 64 X2 Processors, 4GB RAM, 11659 Bogomips Total |
27 |
aemaeth |