README.rst

========
InfluxDB
========

InfluxData is based on the TICK stack, the first open source platform for managing IoT time-series data at scale.

Sample pillars
==============

Single-node influxdb, enabled http frontend and admin web interface:

.. code-block:: yaml

    influxdb:
      server:
        enabled: true
        http:
          enabled: true
          bind:
            address: 0.0.0.0
            port: 8086
        admin:
          enabled: true
          bind:
            address: 0.0.0.0
            port: 8083

Single-node influxdb, SSL for http frontend:

.. code-block:: yaml

    influxdb:
      server:
        enabled: true
        http:
          bind:
            ssl:
              enabled: true
              key_file: /etc/influxdb/ssl/key.pem
              cert_file: /etc/influxdb/ssl/cert.pem

Single-node influxdb where you specify paths for data and metastore directories. Custom
directories are created by this formula:

.. code-block:: yaml

    influxdb:
      server:
        enabled: true
        data:
          dir: '/opt/influxdb/data'
          wal_dir: '/opt/influxdb/wal'
        meta:
          dir: '/opt/influxdb/meta'

InfluxDB server with customized parameters for the data service:

.. code-block:: yaml

    influxdb:
      server:
        enabled: true
        data:
          max_series_per_database: 20000000
          cache_max_memory_size: 524288000
          cache_snapshot_memory_size: 26214400
          cache_snapshot_write_cold_duration: "5m"
          compact_full_write_cold_duration: "2h"2h"
          max_values_per_tag: 5000

Single-node influxdb with an admin user:

.. code-block:: yaml

    influxdb:
      server:
        enabled: true
        http:
          enabled: true
          bind:
            address: 0.0.0.0
            port: 8086
        admin:
          enabled: true
          bind:
            address: 0.0.0.0
            port: 8083
          user:
            enabled: true
            name: root
            password: secret

Single-node influxdb with new users:

.. code-block:: yaml

    influxdb:
      server:
        user:
          user1:
            enabled: true
            admin: true
            name: username1
            password: keepsecret1
          user2:
            enabled: true
            admin: false
            name: username2
            password: keepsecret2

Single-node influxdb with new databases:

.. code-block:: yaml

    influxdb:
      server:
        database:
          mydb1:
            enabled: true
            name: mydb1
          mydb2:
            enabled: true
            name: mydb2

Manage the retention policies for a database:

.. code-block:: yaml

    influxdb:
      server:
        database:
          mydb1:
            enabled: true
            name: mydb1
            retention_policy:
            - name: rp_db1
              duration: 30d
              replication: 1
              is_default: true

Where default values are:

* name = autogen
* duration = INF
* replication = 1
* is_default: false


Here is how to manage grants on database:

.. code-block:: yaml

    influxdb:
      server:
        grant:
          username1_mydb1:
            enabled: true
            user: username1
            database: mydb1
            privilege: all
          username2_mydb1:
            enabled: true
            user: username2
            database: mydb1
            privilege: read
          username2_mydb2:
            enabled: true
            user: username2
            database: mydb2
            privilege: write

InfluxDB relay:

.. code-block:: yaml

    influxdb:
      server:
        enabled: true
        http:
          enabled: true
          output:
            idb01:
              location: http://idb01.local:8086/write
              timeout: 10
            idb02:
              location: http://idb02.local:8086/write
              timeout: 10
        udp:
          enabled: true
          output:
            idb01:
              location: idb01.local:9096
            idb02:
              location: idb02.local:9096

InfluxDB cluster:

.. code-block:: yaml

    influxdb:
      server:
        enabled: true
      meta:
        bind:
          address: 0.0.0.0
          port: 8088
          http_address: 0.0.0.0
          http_port: 8091
      cluster:
        members:
          - host: idb01.local
            port: 8091
          - host: idb02.local
            port: 8091
          - host: idb03.local
            port: 8091

Deploy influxdb apt repository (using linux formula):

.. code-block:: yaml

    linux:
      system:
        os: ubuntu
        dist: xenial
        repo:
          influxdb:
            enabled: true
            source: 'deb https://repos.influxdata.com/${linux:system:os} ${linux:system:dist} stable'
            key_url: 'https://repos.influxdata.com/influxdb.key'

InfluxDB client for configuring databases, users and retention policies:

.. code-block:: yaml

    influxdb:
      client:
        enabled: true
        server:
          protocol: http
          host: 127.0.0.1
          port: 8086
          user: admin
          password: foobar
        user:
          user1:
            enabled: true
            admin: true
            name: username1
        database:
          mydb1:
            enabled: true
            name: mydb1
            retention_policy:
            - name: rp_db1
              duration: 30d
              replication: 1
              is_default: true
        grant:
          username1_mydb1:
            enabled: true
            user: username1
            database: mydb1
            privilege: all

InfluxDB client state's that uses curl can be forced to retry query if curl call fails:

.. code-block:: yaml

    influxdb:
      client:
        enabled: true
        retry:
          count: 3
          delay: 3

Create an continuous queries:

.. code-block:: yaml

    influxdb:
      client:
        database:
          mydb1:
            continuous_query:
              cq_avg_bus_passengers: >-
                SELECT mean("passengers") INTO "transportation"."three_weeks"."average_passengers" FROM "bus_data" GROUP BY time(1h)


Example how to add options to http/udp/data/admin that are not supported by the formula yet:

.. code-block:: yaml

    influxdb:
      server:
        enabled: true
        http:
          enabled: true
          raw_rules:
            - 'max-row-limit = 20000'
        udp:
          enabled: true
          raw_rules:
            - 'max-row-limit = 20000'
        data:
          enabled: true
          raw_rules:
            - 'something_else = 10'
        admin:
          raw_rules:
            - 'something_here = 55'


Rich example for RP and CQ for Telegraf vmstats collected:

.. code-block:: yaml

    influxdb:
      client:
        database:
           vmstats:
             enabled: true
             name: vmstats
             retention_policy:
             - name: a_week
               duration: 10d
               replication: 1
             - name: a_month
               duration: 30d
               replication: 1
             - name: a_quater
               duration: 15w
               replication: 1
               is_default: true
             - name: a_year
               duration: 52w
               replication: 1
             - name: a_decade
               duration: 520w
               replication: 1
             continuous_query:
               cq_ds_all_1m: >-
                 SELECT sum(*) as sum_1m, count(*) as count_1m, median(*) as median_1m, mode(*) as mode_1m, mean(*) as mean_1m, max(*) as max_1m, min(*) as min_1m INTO "vmstats"."a_week".:MEASUREMENT FRO
               cq_ds_all_10m: >-
                 SELECT sum(*) as sum_10m, count(*) as count_10m, median(*) as median_10m, mode(*) as mode_10m, mean(*) as mean_10m, max(*) as max_10m, min(*) as min_10m INTO "vmstats"."a_month".:MEASURE
               cq_ds_all_h: >-
                 SELECT sum(*) as sum_h, count(*) as count_h, median(*) as median_h, mode(*) as mode_h, mean(*) as mean_h, max(*) as max_h, min(*) as min_h INTO "vmstats"."a_month".:MEASUREMENT FROM /.*/
               cq_ds_all_d: >-
                 SELECT sum(*) as sum_d, count(*) as count_d, median(*) as median_d, mode(*) as mode_d, mean(*) as mean_d, max(*) as max_d, min(*) as min_d INTO "vmstats"."a_year".:MEASUREMENT FROM /.*/
               cq_ds_all_w: >-
                 SELECT sum(*) as sum_w, count(*) as count_w, median(*) as median_w, mode(*) as mode_w, mean(*) as mean_w, max(*) as max_w, min(*) as min_w INTO "vmstats"."a_year".:MEASUREMENT FROM /.*/
               cq_ds_all_m: >-
                 SELECT sum(*) as sum_m, count(*) as count_m, median(*) as median_m, mode(*) as mode_m, mean(*) as mean_m, max(*) as max_m, min(*) as min_m INTO "vmstats"."a_decade".:MEASUREMENT FROM /.*
               cq_ds_all_y: >-
                 SELECT sum(*) as sum_y, count(*) as count_y, median(*) as median_y, mode(*) as mode_y, mean(*) as mean_y, max(*) as max_y, min(*) as min_y INTO "vmstats"."a_decade".:MEASUREMENT FROM /.*


Prunning data and data management:

Intended to use in scheduled jobs, executed to maintain data life cycle above retention policy. These states are executed by
``query.sls`` and you are expected to trigger ``sls_id`` individually.

.. code-block:: yaml

    influxdb:
      client:
        database:
          mydb1:
            query:
              drop_measurement_h2o: >-
                DROP MEASUREMENT h2o_quality
              drop_shard_h2o: >-
                DROP SHARD h2o_quality
              drop_series_h2o_feet: >-
                DROP SERIES FROM "h2o_feet"
              drop_series_h2o_feet_loc_smonica: >-
                DROP SERIES FROM "h2o_feet" WHERE "location" = 'santa_monica'
              delete_h2o_quality_rt3: >-
                DELETE FROM "h2o_quality" WHERE "randtag" = '3'
              delete_h2o_quality: >-
                DELETE FROM "h2o_quality"


.. code-block:: shell

    salt \* state.sls_id influxdb_query_delete_h2o_quality influxdb.query


InfluxDB relay with HTTP outputs:

.. code-block:: yaml

    influxdb:
      relay:
        enabled: true
        telemetry:
          enabled: true
          bind:
            address: 127.0.0.1
            port: 9196
        listen:
          http_backend:
            type: http
            bind:
              address: 127.0.0.1
              port: 9096
            output:
              server1:
                location: http://server1:8086/write
                timeout: 20s
                buffer_size_mb: 512
                max_batch_kb: 1024
                max_delay_interval: 30s
              server2:
                location: http://server2:8086/write

Read more
=========

* https://influxdata.com/time-series-platform/influxdb/