How to find the CPU consummed last day on Hadoop without Ambari

Question

I would like to know how to find the CPU consumed last day on my Hadoop cluster without Ambari, because I will need to use a Shell script.

I have ganglia up and running, I can see the graph for the CPU last day on Ambari. If ganglia can give you this result, I suppose it saved the data somewhere? What file? And how can I get it?

Answer 1

This question is already answered here How to get data from ganglia's database

To summarize the answer above

1. The ganglia RRD location is at /var/lib/ganglia/rrds
2. To dump a particular RRD file as plain text use the "rrdtool" command

Both the above info is captured in the question linked above