[GE users] SGE_Hadoop_Error

templedf daniel.templeton at oracle.com
Mon Nov 29 01:00:19 GMT 2010


First, what version of Hadoop are you using?

With just that little bit of information, it sounds like maybe the execd 
went down, causing the load sensor to lose its output stream.

Is this a recurring problem?  If the load sensor fails, the execd should 
just restart it.  When that happens, is it simply failing again?

Daniel

On 11/25/10 4:23 AM, adarsh wrote:
> Hi all,
>
> I am getting confused about getting errors while configuring SGE with Hadoop.
>
> I configured it properly on 4 nodes.
> But on the other day when I tried to configure on new 4 nodes, I face some issues.
>
> my sge_hadoop1.log says :
>
> 11/25/2010 17:44:55|  main|ws34-rak-lin|W|[load_sensor 5997] fflush failed [Broken pipe]
> 11/25/2010 17:44:56|  main|ws34-rak-lin|W|load sensor exited with exit status = 127
>
> And there are no logs on other nodes.
> Please help.
>
> Thanks&  Regards
> Adarsh Sharma
>
> ------------------------------------------------------
> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=298690
>
> To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=300051

To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].



More information about the gridengine-users mailing list