Re: [lxc-devel] Memory Resources

[Daniel Lezcano <daniel.lezcano-GANU6spQydw@public.gmane.org>]

KAMEZAWA Hiroyuki wrote:
> On Fri, 28 Aug 2009 11:32:56 +0200
> Daniel Lezcano <daniel.lezcano-GANU6spQydw@public.gmane.org> wrote:
>   
>>> steps: b, c, d can be done inside lxc tools. step a can't and it is base on
>>> the admin policy.
>>>
>>> I think that the first solution is more automatic and can be done by lxc
>>> tools (maybe command line switch? I can prepare a patch for that.
>>>   
>>>       
>> I do not know smack, what does smack here ? Will this solution avoid the 
>> container to overwrite /proc/meminfo by remounting /proc ?
>>
>>     
> Hmm, how about supporting custom FUSE for private-proc by user land ?
> It seems what virtualized hosts want to see is different from what kernel shows.
> For example, you can show correct "uptime" for virtualized environ.
>   

Yep, interesting and maybe a solution for the application containers.
In the case of the system container, that will be the distro scripts 
which will mount the /proc directory with the usual mount options. The 
objective is to avoid to modify the distro scripts and let them do the 
work, like mounting /proc. If we use a fuse procfs, we will need to 
change these scripts to do the correct mount, no ? (note we have the 
same problem with /cgroup/meminfo solution).

  -- Daniel