You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@storm.apache.org by "rujia (JIRA)" <ji...@apache.org> on 2015/04/29 10:28:05 UTC

[jira] [Updated] (STORM-804) HdfsBolt doesn't work after a period of time which caused by network problems

     [ https://issues.apache.org/jira/browse/STORM-804?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

rujia updated STORM-804:
------------------------
    Description: 
  Two clusters i use in my  test case.
  I submit a topology which write message to hdfs in another cluster,HdfsBolt i use in this topology.It work normally in the beginning, then i intercept the message for a while(30 minites), after that, the bolt write failed all the time.
  I think  HdfsBolt.excute should reconnect and recreate filesystem after it keeping catch exception for a period of time.

  was:
  Two clusters i use in my  test case.
  I submit a topology which write message to hdfs in another cluster,HdfsBolt i use in this topology.It work normally in the beginning, then i intercept the message for a while(5 minites), after that, the bolt write failed all the time.
  I think  HdfsBolt.excute should reconnect and recreate filesystem after it keeping catch exception for a period of time.


> HdfsBolt doesn't work after a period of time which caused by network problems
> -----------------------------------------------------------------------------
>
>                 Key: STORM-804
>                 URL: https://issues.apache.org/jira/browse/STORM-804
>             Project: Apache Storm
>          Issue Type: Bug
>          Components: storm-hdfs
>    Affects Versions: 0.9.3
>            Reporter: rujia
>            Assignee: rujia
>            Priority: Minor
>
>   Two clusters i use in my  test case.
>   I submit a topology which write message to hdfs in another cluster,HdfsBolt i use in this topology.It work normally in the beginning, then i intercept the message for a while(30 minites), after that, the bolt write failed all the time.
>   I think  HdfsBolt.excute should reconnect and recreate filesystem after it keeping catch exception for a period of time.



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