[jira] [Commented] (SAMZA-866) Refactor and fix Container allocation logic.

["Jake Maes (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SAMZA-866?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15127241#comment-15127241 ] 

Jake Maes commented on SAMZA-866:
---------------------------------

Fix:
* Refactor the loop and try-catch to a common run() method in AbstractContainerAllocator
* Change the log type to WARN for the general exception case
* Refactor some duplicate code in ContainerRequestState

> Refactor and fix Container allocation logic. 
> ---------------------------------------------
>
>                 Key: SAMZA-866
>                 URL: https://issues.apache.org/jira/browse/SAMZA-866
>             Project: Samza
>          Issue Type: Bug
>            Reporter: Jake Maes
>            Assignee: Jake Maes
>
> Bug:
> ContainerAllocator and HostAwareContainerAllocator have different try-catch logic in their run() methods. The HostAwareContainerAllocator try-catch is outside the while loop, which would cause it to stop allocating containers in the event of an exception. ContainerAllocator correctly has the try-catch inside the loop 



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