This error will occur if the system doesn't have enough available memory to create a Java virtual machine, that is needed to run a mapreduce job using a RxHadoopMR() compute context. This can happen if you have changed the memory settings for mapreduce for the following:
mapred.child.java.opts
mapred.child.java.ulimit
A workaround for the problem is to reset the setting for those options to the default in Cloudera Manager.
Problem: Hadoop verification script fails: 'Could not reserve enough space for object heap'
Need more help?
Want more options?
Explore subscription benefits, browse training courses, learn how to secure your device, and more.
Communities help you ask and answer questions, give feedback, and hear from experts with rich knowledge.
