Skip to content

Latest commit

 

History

History
130 lines (68 loc) · 3.03 KB

README.md

File metadata and controls

130 lines (68 loc) · 3.03 KB

Template Linux CassandraDB

A Zabbix template for CassandraDB (cluster) monitoring

Based on nagios checks and harisekhon's github repo

Tested on: CentOs 6.x x86_64 Zabbix 2.0.x CassandraDB 1.2.x

next values can be added over JMX util: bash-4.1# java -jar ./cmdline-jmxclient-0.10.3.jar - 127.0.0.1:7199 java.lang:type=Runtime Uptime

and JMX items is: org.apache.cassandra.request:type=* org.apache.cassandra.internal:type=* org.apache.cassandra.db:type=*

Authors

installation

(* install a configure zabbix-agent)

Monitored items

  • Info - Host ID

  • Info - Uptime

  • Info - Native Transport Active

  • Info - Exceptions

  • Info - Thrift Active

  • Info - Gossip Active

  • Info - Heap Memory (Used)

  • Info - Main Status Text

  • Info - Main Status

  • Status - Nodes global status

  • Status - Nodes ALL (text)

  • Status - Nodes Down (count)

  • Status - Nodes Up (count)

  • TPstats - global status

  • TPstats - ReadStage Active

  • TPstats - ReadStage Pending

  • TPstats - ReadStage All time blocked

  • TPstats - ReadStage Blocked

  • TPstats - ReadStage Completed

Trigers

  • CassandraDB - found down nodes on {HOST.NAME} (High)

  • CassandraDB - Gossip state active is not "true" (Not classified)

  • CassandraDB - Heap Memory usage is >85% (Warning)

  • CassandraDB - Heap Memory usage is >95% (Average)

  • CassandraDB - low nodes on {HOST.NAME} (Average)

  • CassandraDB - Main Status is "PROBLEM" on {HOST.NAME} (High)

  • CassandraDB - Nodes global status is not "OK" on {HOST.NAME} (Average)

  • CassandraDB - Thrift state active is not "true" (Not classified)

  • CassandraDB - TPstats is not "OK" on {HOST.NAME} (Average)

  • CassandraDB - Host ID was changed on {HOST.NAME} (Information)

  • CassandraDB - server was rebooted on {HOST.NAME} (Information)

Links