Cassandra Dashboard by ORAMAD

Dashboard

https://github.com/Sarma1807/Prometheus-Grafana-Cassandra
Last updated: 25 days ago

Start with Grafana Cloud and the new FREE tier. Includes 10K series Prometheus or Graphite Metrics and 50gb Loki Logs

Downloads: 30

Reviews: 0

  • CassPromGraf_01.jpg
    CassPromGraf_01.jpg
  • CassPromGraf_02.jpg
    CassPromGraf_02.jpg
  • CassPromGraf_03.jpg
    CassPromGraf_03.jpg
  • CassPromGraf_04.jpg
    CassPromGraf_04.jpg
  • CassPromGraf_05.jpg
    CassPromGraf_05.jpg
  • CassPromGraf_06.jpg
    CassPromGraf_06.jpg
  • CassPromGraf_07.jpg
    CassPromGraf_07.jpg
  • CassPromGraf_08.jpg
    CassPromGraf_08.jpg
  • CassPromGraf_09.jpg
    CassPromGraf_09.jpg
  • CassPromGraf_10.jpg
    CassPromGraf_10.jpg
  • CassPromGraf_11.jpg
    CassPromGraf_11.jpg
  • CassPromGraf_12.jpg
    CassPromGraf_12.jpg
  • CassPromGraf_13.jpg
    CassPromGraf_13.jpg
  • CassPromGraf_14.jpg
    CassPromGraf_14.jpg
  • CassPromGraf_00_Arch.jpg
    CassPromGraf_00_Arch.jpg

This dashboard contains several different sections to highlight the metrics for specific categories.

There are few dedicated sections for Server OS metrics ( OS Details/Storage/CPU/RAM Usage )

All remaining sections are for Cassandra metrics ( Cassandra Nodes Summary/Heap Usage/Cache/CQL/Commit Log Metrics/Clients/Hints/Request Timeouts/Latency Metrics/Size details at Keyspace Level and Table Level )

Dashboard also provides several filters ( cluster/data_center/hostname/keyspace/table ) and makes it very easy to monitor several Apache Cassandra clusters in one place.

Dashboard has been tested and working with Apache Cassandra 3.11.x / Apache Cassandra 4.x & DSE 6.8.x

How Does It Work ?

Apache Cassandra's built in metrics-reporter module is used to extract and publish metrics to a text file on each Cassandra server. Easy to understand bash shell script converts this metrics file to Prometheus metrics format and they are exposed to Prometheus using node_exporter. As you know JMX configuration is complicated, changes to JMX configuration is NOT REQUIRED for this dashboard.

Periodically Prometheus will get the metrics from node_exporter and store them in its own Time-series database.

Grafana is used to read those metrics from Prometheus and display them on an easy to understand dashboard.

Step-by-step configuration and source code has been published at : https://github.com/Sarma1807/Prometheus-Grafana-Cassandra

Dependencies: