Grafana dashboards for measuring MySQL and MongoDB performance using Prometheus

This is a set of Grafana dashboards for database and system monitoring using Prometheus datasource. The dashboards rely on the small patch applied on Grafana (see below).

• Amazon RDS OS metrics (CloudWatch datasource)
• Cross Server Graphs
• Disk Performance
• Disk Space
• MongoDB Cluster Summary
• MongoDB Overview
• MongoDB ReplSet
• MongoDB RocksDB
• MongoDB WiredTiger
• MySQL InnoDB Metrics
• MySQL InnoDB Metrics Advanced
• MySQL MyISAM Metrics
• MySQL Overview
• MySQL Performance Schema
• MySQL Query Response Time
• MySQL Replication
• MySQL Table Statistics
• MySQL TokuDB Graphs
• MySQL User Statistics
• PXC/Galera Cluster Overview
• PXC/Galera Graphs
• Prometheus
• ProxySQL Overview
• Summary Dashboard
• System Overview
• Trends Dashboard

These dashboards are also a part of Percona Monitoring and Management project.

Live demo is available at https://pmmdemo.percona.com/graph/

Setup instructions

Add datasource in Grafana

The datasource should be named Prometheus so it is automatically picked up by the graphs.

Prometheus config

The dashboards use built-in instance label to filter on individual hosts. It is recommended you give the good names to your instances. Here is some example:

scrape_configs:
  - job_name: prometheus
    static_configs:
      - targets: ['localhost:9090']
        labels:
          instance: prometheus

  - job_name: linux
    static_configs:
      - targets: ['192.168.1.7:9100']
        labels:
          instance: db1

  - job_name: mysql
    static_configs:
      - targets: ['192.168.1.7:9104']
        labels:
          instance: db1

How you name jobs is not important. However, "Prometheus" dashboard assumes the job name is prometheus.

Exporter options

Here is the minimal set of options for the exporters:

• node_exporter: -collectors.enabled="diskstats,filefd,filesystem,loadavg,meminfo,netdev,stat,time,uname,vmstat"
• mysqld_exporter: -collect.binlog_size=true -collect.info_schema.processlist=true
• mongodb_exporter: the defaults are fine.

Edit Grafana config

Enable JSON dashboards by uncommenting those lines in grafana.ini:

[dashboards.json]
enabled = true
path = /var/lib/grafana/dashboards

If you wish you may import the individual dashboards via UI and ignore this and the next two steps.

Install dashboards

git clone https://github.com/percona/grafana-dashboards.git
cp -r grafana-dashboards/dashboards /var/lib/grafana/

Restart Grafana

service grafana-server restart

Apply Grafana patch

It is important to apply the following small patch on your Grafana installation in order to use the interval template variable to get the good zoomable graphs. The fix is simply to allow a variable in Step field of graph editor page. For more information, take a look at PR#5839.

Grafana 2.6.0:

sed -i 's/step_input:""/step_input:c.target.step/; s/ HH:MM/ HH:mm/; s/,function(c)/,"templateSrv",function(c,g)/; s/expr:c.target.expr/expr:g.replace(c.target.expr,c.panel.scopedVars)/' /usr/share/grafana/public/app/plugins/datasource/prometheus/query_ctrl.js
sed -i 's/h=a.interval/h=g.replace(a.interval, c.scopedVars)/' /usr/share/grafana/public/app/plugins/datasource/prometheus/datasource.js

Grafana 3.x:

sed -i 's/expr=\(.\)\.replace(\(.\)\.expr,\(.\)\.scopedVars\(.*\)var \(.\)=\(.\)\.interval/expr=\1.replace(\2.expr,\3.scopedVars\4var \5=\1.replace(\6.interval, \3.scopedVars)/' /usr/share/grafana/public/app/plugins/datasource/prometheus/datasource.js
sed -i 's/,range_input/.replace(\/"{\/g,"\\"").replace(\/}"\/g,"\\""),range_input/; s/step_input:""/step_input:this.target.step/' /usr/share/grafana/public/app/plugins/datasource/prometheus/query_ctrl.js

Grafana 4.x (unreleased):

There won't be a need to apply this patch.

Those changes are idempotent and do not break anything.

Update instructions

Simply copy the new dashboards to /var/lib/grafana/dashboards and restart Grafana or re-import them.

Graph samples

Here is some sample graphs.