[jira] [Updated] (HADOOP-9655) IPC Client call to the same host with multi thread takes very long time to report connection time out for many times

["nemon lou (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HADOOP-9655?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

nemon lou updated HADOOP-9655:
------------------------------

    Attachment: HADOOP-9655.patch

This patch use a different object for wait and notify ,so one thread invoking addCall method won't be blocked by another thread calling setupConnection method.
                
> IPC Client call to the same host with multi thread takes very long time to report connection time out for many times 
> ---------------------------------------------------------------------------------------------------------------------
>
>                 Key: HADOOP-9655
>                 URL: https://issues.apache.org/jira/browse/HADOOP-9655
>             Project: Hadoop Common
>          Issue Type: Bug
>          Components: ipc
>    Affects Versions: 2.0.4-alpha
>            Reporter: nemon lou
>         Attachments: HADOOP-9655.patch
>
>
> When one machine power off during running a job ,MRAppMaster find tasks timed out on that host and then call stop container for each container concurrently.
> But the IPC layer did it serially, for each call,the connection time out exception toke a few minutes to raise after 45 times reties. And AM hang for many hours to wait for stopContainer to finish.
> The jstack output file shows that most threads stuck at Connection.addCall waiting for a lock object hold by  Connection.setupIOstreams.
> (The setupIOstreams method run slowlly becauseof connection time out during setupconnection.)

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira