hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Eli Collins (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HADOOP-5464) DFSClient does not treat write timeout of 0 properly
Date Tue, 05 Jul 2011 22:06:16 GMT

     [ https://issues.apache.org/jira/browse/HADOOP-5464?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Eli Collins updated HADOOP-5464:
--------------------------------

     Description: 
{{dfs.datanode.socket.write.timeout}} is used for sockets to and from datanodes. It is 8 minutes
by default. Some users set this to 0, effectively disabling the write timeout (for some specific
reasons). 

When this is set to 0, DFSClient sets the timeout to 5 seconds by mistake while writing to
DataNodes. This is exactly the opposite of real intention of setting it to 0 since 5 seconds
is too short.




  was:

{{dfs.datanode.socket.write.timeout}} is used for sockets to and from datanodes. It is 8 minutes
by default. Some users set this to 0, effectively disabling the write timeout (for some specific
reasons). 

When this is set to 0, DFSClient sets the timeout to 5 seconds by mistake while writing to
DataNodes. This is exactly the opposite of real intention of setting it to 0 since 5 seconds
is too short.




    Release Note: Zero values for dfs.socket.timeout and dfs.datanode.socket.write.timeout
are now respected. Previously zero values for these parameters resulted in a 5 second timeout.
    Hadoop Flags: [Incompatible change, Reviewed]  (was: [Reviewed])

> DFSClient does not treat write timeout of 0 properly
> ----------------------------------------------------
>
>                 Key: HADOOP-5464
>                 URL: https://issues.apache.org/jira/browse/HADOOP-5464
>             Project: Hadoop Common
>          Issue Type: Bug
>    Affects Versions: 0.17.0
>            Reporter: Raghu Angadi
>            Assignee: Raghu Angadi
>             Fix For: 0.21.0
>
>         Attachments: HADOOP-5464.patch
>
>
> {{dfs.datanode.socket.write.timeout}} is used for sockets to and from datanodes. It is
8 minutes by default. Some users set this to 0, effectively disabling the write timeout (for
some specific reasons). 
> When this is set to 0, DFSClient sets the timeout to 5 seconds by mistake while writing
to DataNodes. This is exactly the opposite of real intention of setting it to 0 since 5 seconds
is too short.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message