werty1st / influxdb-grafana

Basic InfluxDB + Grafana Installation

Setup

1. clone repo

   git clone $REPO $FOLDER

2. download and build images

   cd $FOLDER
   docker-compose pull && docker-compose build

3. start containers

   create data dir and change owner

   mkdir -p data/grafana; sudo chown 472:472 data/grafana

   start containers

   docker-compose up -d

4. setup retention policies

   1. connect to InfluxDB shell

      docker-compose exec InfluxDB bash

   2. connect to InfluxDB shell

      influx

   3. issue InfluxDB commands

      USE syslog
      SHOW RETENTION POLICIES
      
      ALTER RETENTION POLICY "autogen" ON "syslog" DURATION 24h REPLICATION 1 DEFAULT
      ALTER RETENTION POLICY "autogen" ON "syslog" DURATION 168h REPLICATION 1 SHARD DURATION 24h DEFAULT
      
      CREATE RETENTION POLICY "two_weeks" ON "syslog" DURATION 2w REPLICATION 1
      CREATE RETENTION POLICY "a_year" ON "syslog" DURATION 52w REPLICATION 1
      
      
      SHOW CONTINUOUS QUERIES
      
      CREATE CONTINUOUS QUERY "cq_p95_5m"  ON "syslog" BEGIN SELECT percentile("duration", 95) AS "duration" INTO "two_weeks"."logstash" FROM "logstash" GROUP BY time(5m),  "engine" END
      DROP CONTINUOUS QUERY "cq_p95_5m" ON "syslog"
      
      
      CREATE CONTINUOUS QUERY "cq_p95_30m" ON "syslog" BEGIN SELECT percentile("duration", 95) AS "duration" INTO "a_year"."logstash" FROM "syslog"."two_weeks"."logstash" GROUP BY time(30m), "engine" END
      DROP CONTINUOUS QUERY "cq_p95_30m" ON "syslog"
      

   4. (optional) remove datapoints inserted before setting up the retention policies

      use syslog
      
      show measurements
      
      delete FROM "logstash" WHERE time <= '2019-07-03T12:00:00.000Z'
      

Usage

Access the Dasboard: http://localhost:3000

• Login: admin
• Password: admin

DEV/DEBUG

> use syslog
Using database syslog
> show measurements
name: measurements
name
----
average_response_1h
logstash
p95_30m
p95_5m
> delete FROM "logstash" WHERE time <= '2019-07-03T12:00:00.000Z'

delete FROM "logstash" WHERE "engine" = 'trending_new'
delete FROM "logstash" WHERE "engine" = 'ts1'
delete FROM "logstash" WHERE "engine" = '%{engine}'

DROP SERIES FROM "logstash" WHERE "engine" = '%{engine}'