influxgraph 1.0.0b1

InfluxDB storage plugin for Graphite-API

InfluxGraph

An InfluxDB 0.9.2 or higher storage plugin for Graphite-API.

This project started as a re-write of graphite_influxdb, now a separate project.

Installation

pip install influxgraph

Use of a local memcached service is highly recommended - see configuration section on how to enable.

Mimimal configuration for Graphite-API is below. See Full Configuration Example for all possible configuration options.

/etc/graphite-api.yaml

finders:
  - influxgraph.InfluxDBFinder

Main features

• InfluxDB Graphite template support - allows for exposure of InfluxDB tagged data as Graphite metrics

• Dynamically calculated group by intervals based on query date/time range - keeps data size tolerable regardless of query date/time range size and speeds up graph generation for large date/time ranges

• Configurable per-query aggregation functions by regular expression pattern

• Configurable per-query retention policies by query date/time range. Use pre-calculated downsampled data in a retention policy for historical data dynamically

• In-memory index for metric path queries

• Multi-fetch enabled - fetch data for multiple metrics with one query to InfluxDB

• Memcached integration

• Python 3 compatibility

Goals

• Backwards compatibility with existing Graphite API clients like Grafana and Graphite installations migrated to InfluxDB backends using Graphite input service with or without Graphite template configuration

• Forwards compatibility with native InfluxDB API input data exposed as Graphite metrics

• Clean, readable code with complete documentation for public endpoints

• Complete code coverage with both unit and integration testing. Code has >90% test coverage and is integration tested against a real InfluxDB service

The two top points provide both

• A backwards compatible migration path for existing Graphite installations to use InfluxDB as a drop-in storage back-end replacement with no API client side changes required, meaning existing Grafana or other dashboards continue to work as-is

• A forwards compatible migration path for native InfluxDB collection agents to expose their data as Graphite API compatible metrics which allows the use of any Graphite API talking tool, the plethora of Graphite API functions, custom functions, multi-series plotting and function support et al

As of this time of writing, no alternatives exist with similar functionality and compatibility.

Dependencies

With the exception of InfluxDB itself, the other dependencies are installed automatically by pip.

• influxdb Python module

• Graphite-API

• python-memcached Python module

• InfluxDB service

Known Limitations

• In memory index can use a lot of memory in InfluxDB installations with a large number of unique metrics (> 1M). Pypy is recommended in that case which allows for a much lower memory footprint compared to the CPython intepreter.

InfluxDB Graphite metric templates

This project can make use of any InfluxDB data and expose them as Graphite API compatible metrics, as well as make use of Graphite metrics added to InfluxDB as-is sans tags.

To make use of tagged InfluxDB data, the plugin needs to know how to parse a Graphite metric path into the tags used by InfluxDB.

The easiest way to do this is to use the Graphite plugin in InfluxDB with a configured template which can be used as-is in InfluxGraph configuration, see Full Configuration Example section for details. This presumes existing metrics collection agents are using the Graphite protocol to write to InfluxDB via its Graphite input service.

If on the other hand, native InfluxDB metrics collection agents like Telegraf are used, that data can too be exposed as Graphite metrics by writing appropriate template(s) in Graphite-API configuration alone.

By default, the storage plugin makes no assumptions that data is tagged, per InfluxDB default Graphite plugin configuration as below:

[[graphite]]
  enabled = true
  # templates = []

Retention policy configuration

Pending implementation of a feature request that will allow InfluxDB to select and/or merge results from multiple retention policies as appropriate, retention policy configuration is needed to support the use-case of down-sampled data being present in non default retention policies.

retention_policies:
    <time interval of query>: <retention policy name>

For example, to make a query with a time interval of ten and thirty minutes use the retention policies named 10min and 30min respectively:

retention_policies:
    600: 10min
    1800: 30min

While not required, retention policy time interval is best kept close to or identical to deltas interval.

Configuration

Minimal Configuration

In graphite-api config file at /etc/graphite-api.yaml:

finders:
  - influxgraph.InfluxDBFinder

The folowing default Graphite-API configuration is used if not provided:

influxdb:
   db: graphite

Full Configuration Example

finders:
  - influxgraph.InfluxDBFinder
influxdb:
    ## InfluxDB configuration
    #
    db: graphite
    host: localhost # (optional)
    port: 8086 # (optional)
    user: root # (optional)
    pass: root # (optional)

    ## Logging configuration
    #
    # Log to file (optional). Default is no finder specific logging.
    log_file: /var/log/graphite_influxdb_finder/graphite_influxdb_finder.log
    # Log file logging level (optional)
    # Values are standard logging levels - `info`, `debug`, `warning`, `critical` et al
    # Default is `info`
    log_level: info

    ## Graphite Template Configuration
    #
    # (Optional) Graphite template configuration
    # One template per line, identical to InfluxDB Graphite input service template configuration
    # See https://github.com/influxdata/influxdb/tree/master/services/graphite for template
    # configuration documentation
    #
    # Note that no special compensation is given to the `field` key if it is used in
    # template configuration and templates with `field` in them should be considered
    # as not fully supported as of yet.
    templates:
      #
      # Template format: [filter] <template> [tag1=value1,tag2=value2]
      #
      ##  Filter, template and extra static tags
      #
      # For a metric path `production.my_host.cpu.cpu0.load` the following template will
      # filter on metrics starting with `environment`,
      # use tags `environment`, `host` and `resource` with measurement name `cpu0.load` and
      # extra static tags `region` and `agent`
      - environment.* environment.host.resource.measurement* region=us-east1,agent=sensu
      #
      ## Template only
      # The following template does not use filter or extra tags.
      # For a metric path `my_host.cpu.cpu0.load` it will use tags `host` and `resource`
      # with measurement name `cpu0.load`
      - host.resource.measurement*
      #
      ## Drop prefix, template with tags after measurement
      # For a metric path `stats.load.my_host.cpu` the following template will use tags
      # `host` and `resource` and remove `stats` prefix from metric paths
      - ..measurement.host.resource
      # A catch-all default template of `measurement*` _should not_ be used -
      # that is the default and would have the same effect as if no template was provided
      #
      ## Examples from InfluxDB Graphite service configuration
      #
      ## filter + template
      # - *.app env.service.resource.measurement
      ## filter + template + extra tag
      # - stats.* .host.measurement* region=us-west,agent=sensu

    ## (Optional) Memcache integration
    #
    memcache:
      host: localhost
      # TTL for /metrics/find endpoint only.
      # TTL for /render endpoint is dynamic and based on data interval.
      # Eg for a 24hr query which would dynamically get a 1min interval, the TTL
      # is 1min.
      ttl: 900 # (optional)
      max_value: 1 # (optional) Memcache (compressed) max value length in MB.

    ## (Optional) Aggregation function configuration
    #
    aggregation_functions:
      # The below four aggregation functions are the
      # defaults used if 'aggregation_functions'
      # configuration is not provided.
      # They will need to be re-added if configuration is provided
      \.min$ : min
      \.max$ : max
      \.last$ : last
      \.sum$ : sum
      # (Optional) Time intervals to use for query time ranges
      # Key is time range of query, value is time delta of query.
      # Eg to use a one second query interval for a query spanning
      # one hour or less use `3600 : 1`
      # Shown below is the default configuration, change/add/remove
      # as necessary.
      deltas:
        # 1 hour -> 1s
        # 3600 : 1
        # 1 day -> 30s
        # 86400 : 30
        # 3 days -> 1min
        259200 : 60
        # 7 days -> 5min
        604800 : 300
        # 14 days -> 10min
        1209600 : 600
        # 28 days -> 15min
        2419200 : 900
        # 2 months -> 30min
        4838400 : 1800
        # 4 months -> 1hour
        9676800 : 3600
        # 12 months -> 3hours
        31536000 : 7200
        # 4 years -> 12hours
        126144000 : 43200

      ## Query Retention Policy configuration
      #
      # (Optional) Retention policies to use for associated time intervals.
      # Key is query time interval in seconds, value the retention policy name a
      # query with the associated time interval, or above, should use.
      #
      # For best performance, retention policies should closely match time interval
      # (delta) configuration values. For example, where delta configuration sets
      # queries 28days and below to use 15min intervals, retention policies would
      # have configuration to use an appropriate retention policy for queries with
      # 15min or above intervals.
      #
      # That said, there is no requirement that the settings be the same.
      #
      # Eg to use a retention policy called `30m` policy for intervals
      # of thirty minutes and above, `10m` for queries with a time
      # interval between thirty to ten minutes and `default` for intervals
      # between ten to five minutes:
      retention_policies:
        1800: 30m
        600: 10m
        300: default

Aggregation function configuration

The graphite-influxdb finder now supports configurable aggregation functions to use for specific metric path patterns. This is the equivalent of storage-aggregation.conf in Graphite’s carbon-cache.

Default aggregation function used is mean, meaning average.

Graphite-influxdb has pre-defined aggregation configuration matching carbon-cache defaults, namely

aggregation_functions:
    \.min$ : min
    \.max$ : max
    \.last$ : last
    \.sum$ : sum

Defaults are overridden if aggregation_functions is configured in graphite-api.yaml as shown in configuration section.

An error will be printed to stderr if a configured aggregation function is not a known valid InfluxDB aggregation method per InfluxDB function list.

Known InfluxDB aggregation functions are defined at graphite_influxdb.constants.INFLUXDB_AGGREGATIONS and can be overriden if necessary.

Note

Please note that when querying multiple series InfluxDB allows only one aggregation function to be used for all series in the query.

In other words, client needs to make sure all series in a wildcard query, for example my_host.cpu.cpu* have the same aggregation function configured.

InfluxGraph will use the first aggregation function configured and log a warning message to that effect if a wildcard query resolves to multiple aggregation functions.

Memcache caching InfluxDB data

Memcache can be used to cache InfluxDB data so the Graphite-API webapp can avoid querying the DB if it does not have to.

TTL configuration for memcache shown above is only for /metrics/find endpoint with /render endpoint TTL being set to the data interval used.

For example, for a query spanning 24hrs, a data interval of 1 min is used by default. TTL for memcache is set to 1 min for that data.

For a query spanning 1 month, a 15min interval is used. TTL is also set to 15min for that data.

Calculated intervals

A data group by interval is automatically calculated depending on the date/time range of the query.

This mirrors what Grafana does when talking directly to InfluxDB.

Overriding the automatically calculated interval is supported via the optional deltas configuration. See Full Configuration Example section for all supported configuration options.

Users that wish to retrieve all data regardless of date/time range are advised to query InfluxDB directly.

Varnish caching InfluxDB API

The following is a sample configuration of Varnish as an HTTP cache in front of InfluxDB’s HTTP API. It uses Varnish’s default TTL of 60 sec for all InfluxDB queries.

The intention is for a local (to InfluxDB) Varnish service to cache frequently accessed data and protect the database from multiple identical requests, for example multiple users viewing the same dashboard.

Graphite-API webapp should use Varnish port to connect to InfluxDB on each node.

Unfortunately, given that clients like Grafana POST requests against the Graphite API, which cannot be cached, using Varnish in front of a Graphite-API webapp would have no effect. Multiple requests for the same dashboard/graph will therefore still hit Graphite-API webapp but with Varnish in front of InfluxDB, the more sensitive DB is spared from duplicated queries.

Substitute the default 8086 backend port with the InfluxDB API port for your installation if needed

backend default {
  .host = "127.0.0.1";
  .port = "8086";
}

sub vcl_recv {
  unset req.http.cookie;
}

Graphite API example configuration

finders:
  - graphite_influxdb.InfluxdbFinder
influxdb:
  db: graphite
  port: <varnish port>

Where <varnish_port> is Varnish’s listening port.

A different HTTP caching service will similarly work just as well.