Re: HDFS file format identification

2016-05-09 Thread Anu Engineer
Unfortunately No. That is a public API. It is annotated with @InterfaceAudience.Public , @InterfaceStability.Stable Here is the Hadoop compatibility info. http://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-common/Compatibility.html Thanks Anu On 5/9/16, 1:15 PM, "Arun Natva"

HDFS file format identification

2016-05-09 Thread Arun Natva
Hi, Right now the HDFS Java api class FileStatus doesn't have the file format as one of the attributes Can this be added as an additional field in FileStatus ? Thanks, Arun Sent from my iPhone - To unsubscribe, e-mail: user-unsu

True multi machine cluster

2016-05-09 Thread Abi
Is Hadoop work on multiple windows machine out of the box. 1. Keyword is "out of box" 2. Same question for yarn and spark ? Please dont ask why we are using windows or other questions. We just are. Would appreciate replies to these questions.

Re: mapper re-run behavior when node dies

2016-05-09 Thread Alvin Chyan
Thanks for the reply, Varun! I quickly got lost in the code since I couldn't figure out which implementation of the interface was being passed into the context. >From your comment, it sounds like ALL the maps that ran on that node need to have their output copied before a map is safe from having t

Re: Filing a JIRA

2016-05-09 Thread Chris Nauroth
Hello John, You should be able to go to https://issues.apache.org/jira and then get an option to create a new issue in any Apache project, including HADOOP, HDFS, YARN or MAPREDUCE, depending on the scope of the issue you want to report. --Chris Nauroth From: John Lilley mailto:john.lil...@red

Filing a JIRA

2016-05-09 Thread John Lilley
I am having trouble filing a bug report. I was trying this: https://issues.apache.org/jira/servicedesk/customer/portal/5/create/27 But this doesn't seem right and refuses my request. Can someone point me to the right place? Thanks John Lilley <> -

Re: Completed Tasks in YARN will not release resources

2016-05-09 Thread Henning Blohm
Did you find YarnChild processes still running? We had problems with Jobs that created non-daemon threads that would prevent the job VM from terminating orderly (I wish the YARN-main of the job would make sure the VM terminates by calling system.exit (or at least make that configurable)). H