[jira] [Commented] (CURATOR-451) Background retry may fall into infinite loop

["Jordan Zimmerman (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CURATOR-451?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16353977#comment-16353977 ] 

Jordan Zimmerman commented on CURATOR-451:
------------------------------------------

Are you seeing an issue in production? Do you have a sample that shows the problem?

> Background retry may fall into infinite loop
> --------------------------------------------
>
>                 Key: CURATOR-451
>                 URL: https://issues.apache.org/jira/browse/CURATOR-451
>             Project: Apache Curator
>          Issue Type: Bug
>          Components: Framework
>    Affects Versions: 4.0.0
>            Reporter: Yongle Zhang
>            Priority: Major
>
> Similar to [Curator-209|https://issues.apache.org/jira/browse/CURATOR-209], addFailedOperation() function in FailedOperationManager.java could fall into infinite loop when there's network disruption. 
> Root cause: when there's an Exception (any type), function addFailedOperation() simply recursively calls itself. And in every recursion, it prints a log.  
>  
> {code:java}
> void addFailedOperation(T details)
> {
>   ...
>   if ( client.getState() == CuratorFrameworkState.STARTED )
>   {
>     log.debug("Details being added to guaranteed operation set: " + details);
>     try
>     {
>       executeGuaranteedOperationInBackground(details);
>     }
>     catch ( Exception e )
>     {
>       ThreadUtils.checkInterrupted(e);
>       addFailedOperation(details);
>     }
>   }
> }
> {code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)