Re: [PATCH 19/30] cr: deal with nsproxy

[Oren Laadan <orenl@cs.columbia.edu>]

Serge E. Hallyn wrote:
> Quoting Alexey Dobriyan (adobriyan@gmail.com):
>> To save nsproxy, or to not save nsproxy?
>>
>> Don't think much, save it.
>>
>> I argue that nsproxy should be removed totally, if someone thinks otherwise. ;-)
> 
> You've got Oren starting to agree with you too.  I personally don't
> much care in principle, and your code looks very nice.

Heh ... as a matter of fact I always agreed with him about that.
(and the irc logs can tell the story :)

In fact, we have much more in agreement than none. That's what
I have been arguing !  Now it's time to settle the disagreements...

Oren.

> 
> The way you do this and the uts patch, though, you (of course) bypass
> the CAP_SYS_ADMIN check in copy_namespaces().  Which is fine for your
> patchset, but a problem if we were to base a compromise patchset on
> your patchset.
> 
> It of course also enforces the 'leakage' checks, which again is
> subject to our whole-container c/r discussion.
> 
> But again, the code is nice, and I see no problems in it.
> 
> -serge
>