> On July 16, 2015, 3:04 p.m., Alexander Rukletsov wrote:
> > src/master/http.cpp, lines 1325-1332
> > <https://reviews.apache.org/r/35983/diff/1/?file=994085#file994085line1325>
> >
> >     Why do we need to recover resources for unreserve?
> 
> Michael Park wrote:
>     If reserved resources are offered, we need to recover the reserved 
> resources before we proceed to updating them to be unreserved.
> 
> Alexander Rukletsov wrote:
>     But you recover all active offers on the slave, not just those with 
> dynamic reservations, or am I missing something?
> 
> Michael Park wrote:
>     I see what you're saying I think. I've updated the code to only rescind 
> the offer if rescinding the offer will contribute to satisfying the request, 
> which I think is more accurate than checking that the offer contains 
> dynamically reserved resources.

I think your change is fine and will do the job.


- Alexander


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/35983/#review91890
-----------------------------------------------------------


On July 28, 2015, 9:06 p.m., Michael Park wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/35983/
> -----------------------------------------------------------
> 
> (Updated July 28, 2015, 9:06 p.m.)
> 
> 
> Review request for mesos, Adam B, Benjamin Hindman, Ben Mahler, Jie Yu, Joris 
> Van Remoortere, and Vinod Kone.
> 
> 
> Bugs: MESOS-2600
>     https://issues.apache.org/jira/browse/MESOS-2600
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> See summary.
> 
> 
> Diffs
> -----
> 
>   src/master/http.cpp 3a1598fad4db03e5f62fd4a6bd26b2bedeee4070 
>   src/master/master.hpp 827d0d599912b2936beb9615610f627f6c9a2d43 
>   src/master/master.cpp 5b5e3c37d4433c8524db267866aebc0a35a181f1 
> 
> Diff: https://reviews.apache.org/r/35983/diff/
> 
> 
> Testing
> -------
> 
> `make check`
> 
> 
> Thanks,
> 
> Michael Park
> 
>

Reply via email to