[ 
https://issues.apache.org/jira/browse/MESOS-2176?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14243302#comment-14243302
 ] 

Benjamin Mahler commented on MESOS-2176:
----------------------------------------

Unit testing refactorings and cleanups committed first:

{noformat}
commit c1421aacd2188f121ef4c34b684671f8a87a16f5
Author: Benjamin Mahler <[email protected]>
Date:   Thu Dec 4 14:27:22 2014 -0800

    Added a TODO in the master for an allocator bug.

    Review: https://reviews.apache.org/r/28812
{noformat}

{noformat}
commit 300327f098045ea818ea4cd6b8a17bc621e210ac
Author: Benjamin Mahler <[email protected]>
Date:   Thu Dec 4 14:40:53 2014 -0800

    Added missing includes to sorter.hpp.

    Review: https://reviews.apache.org/r/28813
{noformat}

{noformat}
commit e72037f3459d40412a5f677f7c3d5edc2bb7a033
Author: Benjamin Mahler <[email protected]>
Date:   Fri Dec 5 12:38:15 2014 -0800

    Updated the allocator interface to enable unit testing.

    Review: https://reviews.apache.org/r/28814
{noformat}

{noformat}
commit 6cf1b016dae90292759dc2dc182fc25a39c58f91
Author: Benjamin Mahler <[email protected]>
Date:   Fri Dec 5 18:38:09 2014 -0800

    Converted an initial DRF integration test to a unit test.

    Review: https://reviews.apache.org/r/28815
{noformat}

{noformat}
commit 3513c21aa0a3c524a59c73f576bd9ec1fbc749ca
Author: Benjamin Mahler <[email protected]>
Date:   Sat Dec 6 18:54:04 2014 -0800

    Moved DRFAllocatorTest.PerSlaveAllocation to a unit test.

    Review: https://reviews.apache.org/r/28816
{noformat}

{noformat}
commit 7cfd1d4611eb4f951e82ac90bc9d9abbecf27277
Author: Benjamin Mahler <[email protected]>
Date:   Sat Dec 6 19:29:04 2014 -0800

    Moved DRFAllocatorTest.SameShareAllocations to a unit test.

    Review: https://reviews.apache.org/r/28817
{noformat}

{noformat}
commit a56c57993e6c847243ee228d131fb48a248aa6a8
Author: Benjamin Mahler <[email protected]>
Date:   Sun Dec 7 16:57:54 2014 -0800

    Moved ReservationAllocatorTest.ReservedResources to a unit test.

    Review: https://reviews.apache.org/r/28818
{noformat}

{noformat}
commit cc70b0eb2bff93fa03e163aaa2362bf276122eb5
Author: Benjamin Mahler <[email protected]>
Date:   Sun Dec 7 17:29:35 2014 -0800

    Moved ReservationAllocatorTest.ResourcesReturned to a unit test.

    Review: https://reviews.apache.org/r/28819
{noformat}

{noformat}
commit feedbfd5c28c129959ba10255bb718daef981467
Author: Benjamin Mahler <[email protected]>
Date:   Sun Dec 7 21:08:32 2014 -0800

    Simplified AllocatorTest.Whitelist by adding a unit test.

    Review: https://reviews.apache.org/r/28820
{noformat}

{noformat}
commit e2239a02c9ea7c541eae5808807c49a2a5676d8a
Author: Benjamin Mahler <[email protected]>
Date:   Wed Dec 3 17:22:56 2014 -0800

    Documented the expected allocator sorting semantics.

    Review: https://reviews.apache.org/r/28821
{noformat}

https://reviews.apache.org/r/28824/ remains

> Hierarchical allocator inconsistently accounts for reserved resources. 
> -----------------------------------------------------------------------
>
>                 Key: MESOS-2176
>                 URL: https://issues.apache.org/jira/browse/MESOS-2176
>             Project: Mesos
>          Issue Type: Bug
>          Components: allocation
>            Reporter: Benjamin Mahler
>            Assignee: Benjamin Mahler
>
> Looking through the allocator code for MESOS-2099, I see an issue with 
> respect to accounting reserved resources in the sorters:
> Within {{HierarchicalAllocatorProcess::allocate}}, only unreserved resources 
> are accounted for in the sorters, whereas everywhere else (add/remove 
> framework, add/remove slave) we account for both reserved and unreserved.
> From git blame, it looks like this issue was introduced over a long course of 
> refactoring and fixes to the allocator. My guess is that this was never 
> caught due to the lack of unit-testability of the allocator (unnecessarily 
> requires a master PID to use an allocator).
> From my understanding, the two levels of the hierarchical sorter should have 
> the following semantics:
> # Level 1 sorts across roles. Only unreserved resources are shared across 
> roles, and therefore the "role sorter" for level 1 should only account for 
> the unreserved resource pool.
> # Level 2 sorts across frameworks, within a role. Both unreserved and 
> reserved resources are shared across frameworks within a role, and therefore 
> the "framework sorters" for level 2 should each account for the reserved 
> resource pool for the role, as well as the unreserved resources _allocated_ 
> inside the role.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to