memtest86+ stopped

memtest86+ stopped in the middle of pass 1 and
and said everything is OK. Should I believe this?
Shouldn't it keep going? BTW, I am running with
the memtest86+ CD removed. OK?

Thanks,
Mike.

Mike -- Email Ignored <> writes:

> memtest86+ stopped in the middle of pass 1 and
> and said everything is OK. Should I believe this?
> Shouldn't it keep going? BTW, I am running with
> the memtest86+ CD removed. OK?

It doesn't need to CD to work. If it stopped partway through the
first pass, something is seriously wrong (it should just keep cranking
until you get tired of it and turn it off).

On Fri, 23 Jan 2009 11:38:32 -0700, Joe Pfeiffer wrote:

> Mike -- Email Ignored <> writes:
>
>> memtest86+ stopped in the middle of pass 1 and and said everything is
>> OK. Should I believe this? Shouldn't it keep going? BTW, I am running
>> with the memtest86+ CD removed. OK?
>
> It doesn't need to CD to work. If it stopped partway through the first
> pass, something is seriously wrong (it should just keep cranking until
> you get tired of it and turn it off).

Pass 1 is the second pass, but I still think its bad.
Mike.

Previously Mike -- Email Ignored <> wrote:
> On Fri, 23 Jan 2009 11:38:32 -0700, Joe Pfeiffer wrote:

>> Mike -- Email Ignored <> writes:
>>
>>> memtest86+ stopped in the middle of pass 1 and and said everything is
>>> OK. Should I believe this? Shouldn't it keep going? BTW, I am running
>>> with the memtest86+ CD removed. OK?
>>
>> It doesn't need to CD to work. If it stopped partway through the first
>> pass, something is seriously wrong (it should just keep cranking until
>> you get tired of it and turn it off).

> Pass 1 is the second pass, but I still think its bad.
> Mike.

I agree. And I think that you should run at least 5 passes, better more.
I once had a bad bit in one Infineon module that took about 20 passes
to be discoverd.

Arno

Previously Joe Pfeiffer <> wrote:
> Arno Wagner <> writes:

>> Previously Mike -- Email Ignored <> wrote:
>>> On Fri, 23 Jan 2009 11:38:32 -0700, Joe Pfeiffer wrote:
>>
>>>> Mike -- Email Ignored <> writes:
>>>>
>>>>> memtest86+ stopped in the middle of pass 1 and and said everything is
>>>>> OK. Should I believe this? Shouldn't it keep going? BTW, I am running
>>>>> with the memtest86+ CD removed. OK?
>>>>
>>>> It doesn't need to CD to work. If it stopped partway through the first
>>>> pass, something is seriously wrong (it should just keep cranking until
>>>> you get tired of it and turn it off).
>>
>>> Pass 1 is the second pass, but I still think its bad.
>>> Mike.
>>
>> I agree. And I think that you should run at least 5 passes, better more.
>> I once had a bad bit in one Infineon module that took about 20 passes
>> to be discoverd.

> The statement that it just stopped rather accumulating errors makes me
> wonder if the problem might be CPU rather than memory, though...

Possible. It does use a tiny amount of memory for itself, if I
remember correctly, but that is very unlikely for that to be hit.

Should not be caches though, they are tested also.

Would be interesting to know what exactly the error was.
Aslo a second run should have different results. If not,
a CPU problem seems very likely.

Arno

On Fri, 23 Jan 2009 17:53:52 +0000, Mike -- Email Ignored wrote:

> memtest86+ stopped in the middle of pass 1 and and said everything is
> OK. Should I believe this? Shouldn't it keep going? BTW, I am running
> with the memtest86+ CD removed. OK?
>
> Thanks,
> Mike.

I just ran memtest86+ for more than 20 passes on
the same box with no problem. I am beginning to
wonder if I misinterpreted something I saw.

Mike.

Arno Wagner <> writes:

> Previously Joe Pfeiffer <> wrote:
>> Arno Wagner <> writes:
>
>>> Previously Mike -- Email Ignored <> wrote:
>>>> On Fri, 23 Jan 2009 11:38:32 -0700, Joe Pfeiffer wrote:
>>>
>>>>> Mike -- Email Ignored <> writes:
>>>>>
>>>>>> memtest86+ stopped in the middle of pass 1 and and said everything is
>>>>>> OK. Should I believe this? Shouldn't it keep going? BTW, I am running
>>>>>> with the memtest86+ CD removed. OK?
>>>>>
>>>>> It doesn't need to CD to work. If it stopped partway through the first
>>>>> pass, something is seriously wrong (it should just keep cranking until
>>>>> you get tired of it and turn it off).
>>>
>>>> Pass 1 is the second pass, but I still think its bad.
>>>> Mike.
>>>
>>> I agree. And I think that you should run at least 5 passes, better more.
>>> I once had a bad bit in one Infineon module that took about 20 passes
>>> to be discoverd.
>
>> The statement that it just stopped rather accumulating errors makes me
>> wonder if the problem might be CPU rather than memory, though...
>
> Possible. It does use a tiny amount of memory for itself, if I
> remember correctly, but that is very unlikely for that to be hit.
>
> Should not be caches though, they are tested also.
>
> Would be interesting to know what exactly the error was.
> Aslo a second run should have different results. If not,
> a CPU problem seems very likely.

Well, the OP said there was no error. It just stopped.