[
https://issues.apache.org/jira/browse/TAJO-296?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
hyoungjunkim updated TAJO-296:
------------------------------
Attachment: TAJO-296.patch
I changed jetty's SessionIdManager to HashSessionIdManager.
(http://docs.codehaus.org/display/JETTY/Connectors+slow+to+startup)
> Late registration of Tajo workers
> ---------------------------------
>
> Key: TAJO-296
> URL: https://issues.apache.org/jira/browse/TAJO-296
> Project: Tajo
> Issue Type: Bug
> Components: resource manager
> Reporter: Jihoon Son
> Fix For: 0.8-incubating
>
> Attachments: TAJO-296.patch
>
>
> The worker registration takes too much time.
> When I tested on an in-house cluster which consists of one master and 32
> workers, 31 workers are registered in about 2 minutes and one worker is
> failed to register.
> My guess from the log is that the failed worker does not receive any start
> messages.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
