[jira] [Commented] (EAGLE-848) eagle should log proper error instead of dying

["Hao Chen (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/EAGLE-848?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15793917#comment-15793917 ] 

Hao Chen commented on EAGLE-848:
--------------------------------

[~jhsenjaliya] are you starting in local mode? If yes, it's because **Storm Local Cluster** shutdown (exit) the current process when any fetal failure. 

> eagle should log proper error instead of dying
> ----------------------------------------------
>
>                 Key: EAGLE-848
>                 URL: https://issues.apache.org/jira/browse/EAGLE-848
>             Project: Eagle
>          Issue Type: Improvement
>          Components: Alert Engine
>    Affects Versions: v0.5.0
>            Reporter: Jayesh
>             Fix For: v0.5.0
>
>
> on several occasions, I found eagle server dies because the exception is not handled, we should gather here such places, and put up proper exception handling to log the error that tells users of why exactly something didnt work.
> here is 1 example to begin with.
> eagle-server died because the topic mentioned to start application didnt exist in zookeeper.
> ! Causing: java.lang.RuntimeException: java.lang.RuntimeException: org.apache.zookeeper.KeeperException$NoNodeException: KeeperErrorCode = NoNode for /brokers/topics/hadoop_jmx_metric/partitions
> it would be even better to provide dropdown list of suggestion to choose topic from.



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