check_hadoop-dfs.sh

Project Notes

Description Highly work in progress since I'm currently occupied with kicking off a large Hadoop cluster and the checks are written en passant. Please note that there's a small shell snippet that needs to be sudo'd by the Nagios user to get the actual dfs statistics. This is because I didn't want to give hadoop related permissions to the Nagios user. Put the small snippet attached below in a directory of your choice (configurable via -s/--path-sh), but name it get-dfsreport.sh and make it read, write and accessible by root only. Then enable the Nagios user via /etc/sudoers (or better visudo) to run the script. Version Version 1.0, 2009, Mike Adolphs (http://www.matejunkie.com/) -h/--help Output check_hadoop-dfs.sh is a Nagios plugin to check the status of HDFS, Hadoop's underlying, redundant, distributed file system. check_hadoop-dfs.sh -s /usr/local/sbin -w 10 -c 5 Options: -s|--path-sh) Path to the shell script that is mentioned in the 1. Default is: /usr/local/sbin -w|--warning) Defines the warning level for available datanodes. Default is: off -c|--critical) Defines the critical level for available datanodes. Default is: off Output Example user@host: ~ $ ./check_hadoop-dfs.sh -s /var/nagios/home/bin OK - Datanodes up and running: 50, DFS total: 20147365 MB, DFS used: 0 MB (0%) | 'datanodes_available'=50 'dfs_total'=20147365 'dfs_used'=0 user@host: ~ $ ./check_hadoop-dfs.sh -s /var/nagios/home/bin -w 40 -c 30 OK - Datanodes up and running: 50, DFS total: 20147365 MB, DFS used: 0 MB (0%) | 'datanodes_available'=50 'dfs_total'=20147365 'dfs_used'=0 user@host: ~ $ ./check_hadoop-dfs.sh -s /var/nagios/home/bin -w 60 -c 40 WARNING - Datanodes up and running: 50, DFS total: 20147365 MB, DFS used: 0 MB (0%) | 'datanodes_available'=50 'dfs_total'=20147365 'dfs_used'=0 user@host: ~ $ ./check_hadoop-dfs.sh -s /var/nagios/home/bin -w 70-c 60 CRITICAL - Datanodes up and running: 50, DFS total: 20147365 MB, DFS used: 0 MB (0%) | 'datanodes_available'=50 'dfs_total'=20147365 'dfs_used'=0 user@host: ~ $ ./check_hadoop-dfs.sh -s /var/nagios/home/bin -w 20 -c 40 Please adjust your warning/critical thresholds. The warning must be higher than the critical level! user@host: ~ $ ./check_hadoop-dfs.sh -s /var/nagios/home/bin -w 30 Please also set a critical value when you want to use warning/critical thresholds! user@host: ~ $ ./check_hadoop-dfs.sh -s /var/nagios/home/bin -c 30 Please also set a warning value when you want to use warning/critical thresholds!