Re: just a dump

[Hans de Bruin <bruinjm@xs4all.nl>]

Lucas Meneghel Rodrigues wrote:
> On Sat, 2009-05-16 at 10:38 +0200, Hans de Bruin wrote:
>> Marcelo Tosatti wrote:
>>> On Wed, May 13, 2009 at 12:20:26AM +0200, Hans de Bruin wrote:
>>>> Hans de Bruin wrote:
>>>>> Staring to vms simultaneously end in crash
>>>>>
>>>>> linux 30-rc5
>>>>> kvm-qemu kvm-85-378-g143eb2b
>>>>> proc AMD dualcore
>>>>>
>>>>> vm's like:
>>>>>
>>>>> #!/bin/sh
>>>>> n=10
>>>>> cdrom=/iso/server2008x64.iso
>>>>> drive=file=/kvm/disks/vm$n
>>>>> mem=1024
>>>>> cpu=qemu64
>>>>> vga=std
>>>>> mac=52:54:00:12:34:$n
>>>>> bridge=br1
>>>>>
>>>>> qemu-system-x86_64 -cdrom $cdrom -drive $drive -m $mem -cpu $cpu -vga  
>>>>> $vga -net nic,macaddr=$mac -net tap,script=/etc/qemu/$bridge
>>>>>
>>>>>
>>>> another dmesg:
>>> Hans,
>>>
>>> The oopses below point to the possibility of a hardware problem, 
>>> similar to:
>>>
>>> https://bugzilla.redhat.com/show_bug.cgi?id=480779
>>>
>>> Can you please rule it out with memtest86?
>> I ran memtest for 11 hours and it completed 4.7 passes with no problems.
>> But then memtest is about cpu and memmory interaction. If the problem is 
>> disk related there is also disk/chipset/dma and memmory interaction. I 
>> could degrade my system by turning dma on disk io off, or i could have a 
>> closer look at kvm-autotest.
> 
> Hans:
> 
> There is a memory test designed to stress disk/chipset/dma interaction.
> I made an implementation of it on autotest, the test module is called
> dma_memtest:
> 
> http://autotest.kernel.org/browser/trunk/client/tests/dma_memtest/dma_memtest.py

[09:09:47 INFO ] Test finished after 1 iterations.
Memory test passed.
[09:09:48 DEBUG] Running 'grep MemTotal /proc/meminfo'
[09:09:48 DEBUG] Running 'rpm -qa'
[09:09:49 INFO ]                GOOD    dma_memtest     dma_memtest 
timestamp=1243408189    localtime=May 27 09:09:49       completed 
successfully
[09:09:49 DEBUG] Persistent state variable __group_level now set to 1
[09:09:49 INFO ]        END GOOD        dma_memtest     dma_memtest 
timestamp=1243408189    localtime=May 27 09:09:49
[09:09:49 DEBUG] Dropping caches
[09:09:49 DEBUG] Running 'sync'
[09:09:51 DEBUG] Running 'sync'
[09:09:51 DEBUG] Running 'echo 3 > /proc/sys/vm/drop_caches'
[09:09:52 DEBUG] Persistent state variable __group_level now set to 0
[09:09:52 INFO ] END GOOD       ----    ----    timestamp=1243408192 
localtime=May 27 09:09:52

Well that looks good. The web page talks about forcing the system to 
swap. That never happend swap usage is still 0 bytes. I installed 
autotest on the same lv (3 disk stripe) as the vmdisks.

-- 
Hans