You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@eagle.apache.org by "hemanth dendukuri (JIRA)" <ji...@apache.org> on 2016/02/22 21:49:18 UTC

[jira] [Reopened] (EAGLE-4) configure parallelism with bolt/spout friendly name

     [ https://issues.apache.org/jira/browse/EAGLE-4?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

hemanth dendukuri reopened EAGLE-4:
-----------------------------------

re-opened to update "fix Version".

> configure parallelism with bolt/spout friendly name
> ---------------------------------------------------
>
>                 Key: EAGLE-4
>                 URL: https://issues.apache.org/jira/browse/EAGLE-4
>             Project: Eagle
>          Issue Type: Bug
>         Environment: production
>            Reporter: Edward Zhang
>            Assignee: Zhao, Qingwen
>            Priority: Minor
>             Fix For: 0.3.0
>
>
> In order to define parallelism for each spout/bolt, we should use friendly name. Today normally Eagle API does not require bolt name is provided, but to define parallelism, we have to use a friendly name.
> An alternative is to use rebalance command like bin/storm rebalance apollo-phx-hdfsAuditLog-topology -n 10 -e kafkaMsgConsumer=1 -e FileSensitivityDataJoinExecutor_2=1 -e JavaStormExecutorForAlertWrapper_5=1 -e hdfsAuditLogAlertExecutor_0=1
> But the best is to provide a way of defining parallelism in configuration file
> "parallelismConfig" : {
>       "kafkaMsgConsumer" : 2,
>       "FileSensitivityDataJoinExecutor_2" : 2,
>       "hdfsAuditLogAlertExecutor*" : 1
>     }



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