yes i did recompile. But i seem to face the same problem. I am running the map reduce with a custom input format. I am not sure if there is some change in the API to get the splits correct.


On Tue, Jul 16, 2013 at 6:24 AM, 闫昆 <> wrote:
I think you should recompile the program after run the program

2013/7/13 Anit Alexander <>

I am encountering a problem in cdh4 environment. 
I can successfully run the map reduce job in the hadoop cluster. But when i migrated the same map reduce to my cdh4 environment it creates an error stating that it cannot read the next block(each block is 64 mb). Why is that so?

Hadoop environment: hadoop 1.0.3
java version 1.6

chd4 environment: CDH4.2.0
java version 1.6

Anit Alexander