Database Reference
In-Depth Information
Figure 7-17.
Ganglia node display
While this multitude of resource graphs may be bewildering at first, it does present a wide range of possible
features that Ganglia can be used to monitor. It should be obvious that Ganglia provides suitable functionality to
monitor a Hadoop cluster. Check websites like
ganglia.sourceforge.net
for further information. Also, try installing
Ganglia yourself and compare the graphs that it produces by default against your own Hadoop cluster.
This chapter also presents the Nagios monitoring system. Nagios extends the possibilities for system monitoring
by providing the ability to create alerts based on the problems that it finds.
Nagios
Nagios is an open-source cluster monitoring system that is available from Nagios Enterprises
(
www.nagios.org
)
.
While systems like Ganglia offer a wide range of monitoring graph options, the Nagios monitoring system provides
the ability to monitor cluster server attributes and create alerts when problems occur. Nagios has a similar structure to
Ganglia, in that it has a Nagios Master server and client (nrpe) programs. In this section, I show how it can be sourced,
installed, configured, and used to monitor alerts.
Search WWH ::
Custom Search