-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/38110/#review105318
-----------------------------------------------------------



src/master/quota_handler.cpp (line 159)
<https://reviews.apache.org/r/38110/#comment163843>

    Can you append "via `flatten`" at the end.  Just to be clear how you're 
removing the role.



src/master/quota_handler.cpp (lines 166 - 168)
<https://reviews.apache.org/r/38110/#comment163847>

    I checked the codepaths for `Operation::RESERVE` and apparently there's 
nothing preventing you from reserving revocable resources.  (Unless I'm not 
seeing it.)
    
    I think we shouldn't allow this.  If we did, we may end up double counting 
(or worse) the same resources.  For example, if the `ResourceEstimator` always 
returned the total CPU on the agent, we would see that agent as having 2x CPU 
(half revocable).


- Joseph Wu


On Nov. 5, 2015, 11:29 a.m., Alexander Rukletsov wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/38110/
> -----------------------------------------------------------
> 
> (Updated Nov. 5, 2015, 11:29 a.m.)
> 
> 
> Review request for mesos, Bernd Mathiske, Joerg Schad, Joris Van Remoortere, 
> Joseph Wu, and Michael Park.
> 
> 
> Bugs: MESOS-3074
>     https://issues.apache.org/jira/browse/MESOS-3074
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Performs a check whether a quota request is reasonable and can be satisfied 
> at the moment. A precise answer is impossible here, but a sanity check is 
> still helpful, because it allows us to filter knowingly unsatisfiable 
> requests.
> 
> 
> Diffs
> -----
> 
>   src/master/master.hpp 3aa7017287506938d77bbfb1e06f510101009ee3 
>   src/master/quota_handler.cpp PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/38110/diff/
> 
> 
> Testing
> -------
> 
> make check (Mac OS X 10.10.4)
> 
> 
> Thanks,
> 
> Alexander Rukletsov
> 
>

Reply via email to