Database Reference
In-Depth Information
Summary
In this chapter, we learned various details of Impala troubleshooting through Cloudera
Manager: log analysis, checking events, console output, and so on. We have seen
how Cloudera Manager can be very useful to troubleshoot various problems in Impala
as well as how you can look for potential performance-specific issues in logs. A manu-
al study of the log is very important to learn more about Impala execution and once
you understood it very well, you can troubleshoot the problem just by revisiting the
log. There are several other factors that can be considered as potential problems that
impact Impala performance. Sometimes, the Hadoop cluster itself is very busy per-
forming several
MapReduce
jobs submitted by other issues. This can consume signi-
ficant resources from nodes in the Hadoop cluster and ultimately cause problems in
Impala execution. Networking issues, such as a congested network, slow performing
network cards, and network limitations of any kind could also cause potential perform-
ance issues with Impala. In most of these situations, cluster and logs analysis is one
of the best options to find the root cause. Then, apply the specific information that you
learned in this chapter to solve your problem.
extend Impala to the next level and make it much more useful. In the next chapter, we
will cover a few advanced topics such as HBase integration and HDFS file formats to
increase your knowledge of Impala.
