You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@storm.apache.org by "Jungtaek Lim (JIRA)" <ji...@apache.org> on 2015/11/23 04:42:10 UTC
[jira] [Updated] (STORM-1224) ShellBolt only increases memory usage
when subprocess doesn't send ack / fail
[ https://issues.apache.org/jira/browse/STORM-1224?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jungtaek Lim updated STORM-1224:
--------------------------------
Description:
In order to handle ack / fail easily, ShellBolt stores generated id and tuple to Map, and remove it when subprocess passes ack or fail to that id.
In other word, if users doesn't let subprocess pass ack or fail, _input Map only increases and never decreases. It would make memory issue when each tuple is big or there're a lot of tuples flowing.
was:
In order to handle ack / fail easily, ShellBolt stores generated id and tuple to Map, and remove it when subprocess passes ack or fail to that id.
In other word, if users doesn't make subprocess pass ack or fail, _input Map only increases and never decreases. It would make memory issue when each tuple is big or there're a lot of tuples flowing.
> ShellBolt only increases memory usage when subprocess doesn't send ack / fail
> -----------------------------------------------------------------------------
>
> Key: STORM-1224
> URL: https://issues.apache.org/jira/browse/STORM-1224
> Project: Apache Storm
> Issue Type: Bug
> Components: storm-core, storm-multilang
> Reporter: Jungtaek Lim
>
> In order to handle ack / fail easily, ShellBolt stores generated id and tuple to Map, and remove it when subprocess passes ack or fail to that id.
> In other word, if users doesn't let subprocess pass ack or fail, _input Map only increases and never decreases. It would make memory issue when each tuple is big or there're a lot of tuples flowing.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)