Usage
=====

Configuration
-------------

All remote-connection details are configured via environment-variables. The
only "core" environment variable is ``PGFLUX_POSTGRES_DSN``. This is used to
open a connection to the PostgreSQL cluster to fetch the statistics. Example::

    PGFLUX_POSTGRES_DSN=postgresql://postgres:mys3cr37@localhost/postgres

Quickstart
----------

With only the ``PGFLUX_POSTGRES_DSN`` value set you can run::

    pgflux --all

This runs all available queries against the given database cluster and writes
the result to ``stdout``.


Outputs
-------

``pgflux`` supports different outputs. By default it bundles with ``stdout``,
``http`` and ``https``. To see the available outputs, run::

    pgflux --list-outputs

This will also include any environment variables needed by that output.

The bundled outputs are:

``stdout``
    Write data to standard output

``http`` / ``https``
    Write the data directly into InfluxDB using HTTP(s) POST requests. This
    requires additional environment variables (see the output of
    ``--list-outputs``).


Queries
-------

``pgflux`` comes with a set of bundled queries. To select queries, specify them
one by one on the command-line. For example::

    pgflux db:disk_io cluster:query_ages

To list all available queries, run::

    pgflux --list-queries

Some queries may appear multiple times in case the implementation changed
between PostgreSQL versions.

The queries fall into two categories, identified by their prefix:

db (Database-Local)
    These queries are relative to the current connection. As such, to fetch
    these statistics, a new connection to the database needs to be opened.

cluster
    Tese queries are independent of the current connection and work for the
    whole database cluster.

Additionally, each query has a *minimum* version attached representing the
PostgreSQL version where this query starts being supported.