Re: Best practices configuring libraries on the backend.

2012-03-28 Thread Harsh J
George, This ought to work. Did you restart all your TTs to have it set into effect? Also, the right way to do this across Hadoop (in 1.0/cdh3/whatever) is to add into your hadoop-env.sh: JAVA_LIBRARY_PATH=/path/to/your/libs:$JAVA_LIBRARY_PATH This way you do not stand to lose Hadoop's native l

Re: Send a map to all nodes

2012-03-28 Thread Harsh J
Luiz, Though it is possible to 'hint' this by tweaking the InputSplits passed from the job, the default schedulers of Hadoop do not make any such guarantees and hence this isn't possible unless you write your own complete scheduler, an exercise that wouldn't suit production deployments unless you

Possible to poll JobTracker for information from any language?

2012-03-28 Thread Ryan Cole
Hello, I'm interested in writing a library, to be used with Node.js, that can ask the JobTracker for information about jobs. I see that this is possible using the Java API, with the JobClient interface [1]. I also saw that on the wiki, it mentions that clients can poll the JobTracker for informati

Send a map to all nodes

2012-03-28 Thread Luiz Carlos Muniz
Hi, Is there any way to ensure the execution of a map on all nodes of a clusterin a way that each node run the map once and only once. That is, I would use Hadoop to execute a method on all nodes in the cluster. Without the possibility that the method execute twice in the same node even if another

Re: Best practices configuring libraries on the backend.

2012-03-28 Thread George Datskos
Dmitriy I've tested it on hadoop 1.0.0 and 1.0.1. (I don't know which version cdh3u3 is based off of) In hadoop-env.sh if I set HADOOP_TASKTRACKER_OPTS="-Djava.library.path=/usr/blah" the TaskTracker sees that option. Then it gets passed along to all M/R child tasks on that node. Can you

Re: Best practices configuring libraries on the backend.

2012-03-28 Thread Dmitriy Lyubimov
Hm. doesn't seem to work for me (with cdh3u3) I defined export HADOOP_TASKTRACKER_OPTS="-Djava.library.path=/usr/" and it doesn't seem to work (as opposed to when i set is with property mapred.child.java.opts on the data node). Still puzzling. On Tue, Mar 27, 2012 at 7:17 PM, George Datsko

Re: Best practices configuring libraries on the backend.

2012-03-28 Thread Bharath Mundlapudi
Dmitriy, You can set for map or reduce tasks. Please refer this link: http://hadoop.apache.org/common/docs/r1.0.1/mapred_tutorial.html#Task+Execution+%26+Environment mapred.map.child.java.opts -Xmx512M -Djava.library.path=/home/mycompany/lib -verbose:gc -Xloggc:/tmp/@taskid@.gc