Where did all my memory go?

Charles Lopes tjarls at iee.lu
Fri Feb 24 00:07:27 UTC 2006


Arjan van de Ven wrote:
> On Thu, 2006-02-23 at 09:16 -0500, Ivan Gyurdiev wrote:
>   
>>> And how is anybody here going to be able to debug or even only guess
>>> what is wrong when using a kernel like that ?
>>>   
>>>       
>> Well, I would assume, by having a better idea about how memory 
>> allocations work, and how they can be tracked, people could hint me as 
>> to how to figure out what's eating all the memory (since the top command 
>> isn't helping). Maybe three's a command I don't know about, or /proc 
>> statistics that could explain what's happening. If I knew how to debug 
>> the problem I wouldn't be asking the question, would I?
>>
>> I guess I should cc Steve Grubb on this, since the kernel does include 
>> lspp patches.
>>     
>
> also ask nvidia and madwifi for help first I suppose ;)
>
>   
Sure he's got these infamous modules. But he's not asking for someone to 
solve his problem but for information on how he could do it. Or does it 
mean that, just because he took the risk to use these modules, he should 
be denied the information he seeks. So please, would anyone be so kind 
as to explain how to track such leaks. I, for one, would be interested 
in knowing how to track them too. After three days, I get this:
3037932 3036704  99%    0.04K  33021       92    132084K biovec-1
3037920 3036816  99%    0.12K 101264       30    405056K bio
I guess my sin is to have LVM on the top of MD RAID1.




More information about the devel mailing list