curator-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CURATOR-116) Ordering of delayed distributed queue is inconsistent
Date Wed, 25 Jun 2014 13:42:25 GMT

    [ https://issues.apache.org/jira/browse/CURATOR-116?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14043449#comment-14043449
] 

ASF GitHub Bot commented on CURATOR-116:
----------------------------------------

Github user Randgalt commented on a diff in the pull request:

    https://github.com/apache/curator/pull/13#discussion_r14185554
  
    --- Diff: curator-recipes/src/main/java/org/apache/curator/framework/recipes/queue/DistributedDelayQueue.java
---
    @@ -77,14 +77,22 @@
                 finalFlushMs
             )
             {
    +            @Override
                 protected long getDelay(String itemNode)
                 {
    +                return getDelay(itemNode, System.currentTimeMillis());
    +            }
    +            
    +            protected long getDelay(String itemNode, long sortTime)
    --- End diff --
    
    This can be private, right?


> Ordering of delayed distributed queue is inconsistent
> -----------------------------------------------------
>
>                 Key: CURATOR-116
>                 URL: https://issues.apache.org/jira/browse/CURATOR-116
>             Project: Apache Curator
>          Issue Type: Bug
>          Components: Recipes
>    Affects Versions: 2.5.0
>            Reporter: Cameron McKenzie
>            Priority: Minor
>              Labels: DelayedDistributedQueue, recipes
>
> The ordering in which elements in the delayed distributed queue are processed is inconsistent.
Only elements that have reached their expiry time are processed, but their actual order is
not deterministic due to the logic used for sorting. The current time is used during the sorting
process, so if an element becomes ready for processing half way through the sort, the order
is not deterministic.
> The current time should be determined at the start of the sort process, and this point
in time used for all comparisons.
> I believe this is the root cause for CURATOR-115 also, but I cannot reproduce CURATOR-115,
so I have raised this as a separate defect.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message