Hi Jyoti;
I assume this is the log of master vertex. It seems like master can not
reach a worker for some reason. Did you also check the worker vertex's
log? Maybe you can share it too.
Sertug
On 20-01-2014 09:22, Jyoti Yadav wrote:
*h.master.MasterThread: masterThread: Master algorithm
Hello Artie;
I also had such problems when working on large datasets. It probably
needs a better hadoop/mapreduce configuration. For example did you try
increasing the memory of task tracker child processes in mapred-site.xml
such as:
property
namemapred.child.java.opts/name
Hi all;
I have a 10 machine cluster; 8 Tasktrackers, 1 namenode and 1
Jobtracker. All of them have 8GB memory.
When I run a giraph job with 16 workers on a hbase table with 15
millions of rows (12GB of data) my job does not finish due to 2
(sometimes 3) workers stuck at superstep -1. Below