# Final Aggregator Plugin

The final aggregator emits the last metric of a contiguous series.  A
contiguous series is defined as a series which receives updates within the
time period in `series_timeout`. The contiguous series may be longer than the
time interval defined by `period`.

This is useful for getting the final value for data sources that produce
discrete time series such as procstat, cgroup, kubernetes etc.

When a series has not been updated within the time defined in
`series_timeout`, the last metric is emitted with the `_final` appended.

## Global configuration options <!-- @/docs/includes/plugin_config.md -->

In addition to the plugin-specific configuration settings, plugins support
additional global and plugin configuration settings. These settings are used to
modify metrics, tags, and field or create aliases and configure ordering, etc.
See the [CONFIGURATION.md][CONFIGURATION.md] for more details.

[CONFIGURATION.md]: ../../../docs/CONFIGURATION.md#plugins

## Configuration

```toml @sample.conf
# Report the final metric of a series
[[aggregators.final]]
  ## The period on which to flush & clear the aggregator.
  period = "30s"
  ## If true, the original metric will be dropped by the
  ## aggregator and will not get sent to the output plugins.
  # drop_original = false

  ## The time that a series is not updated until considering it final.
  # series_timeout = "5m"

  ## Output strategy, supported values:
  ##   timeout  -- output a metric if no new input arrived for `series_timeout`;
  ##               useful for filling gaps in input data
  ##   periodic -- output the last received metric every `period`; useful to
  ##               downsample the input data
  # output_strategy = "timeout"
```

### Output strategy

By default (`output_strategy = "timeout"`) the plugin will only emit a metric
for the period if the last received one is older than the series_timeout. This
will not guarantee a regular output of a `final` metric e.g. if the
series-timeout is a multiple of the gathering interval for an input. In this
case metric sporadically arrive in the timeout phase of the period and emitting
the `final` metric is suppressed.
This can be helpful to fill in gaps in the data if no input arrived in time.

Contrary to this, `output_strategy = "periodic"` will always output a `final`
metric at the end of the period irrespectively of when the last metric arrived,
the `series_timeout` is ignored.
This is helpful if you for example want to downsample input data arriving at a
high rate and require a periodic output of the `final` metric.

## Metrics

Measurement and tags are unchanged, fields are emitted with the suffix
`_final`.

## Example Output

```text
counter,host=bar i_final=3,j_final=6 1554281635115090133
counter,host=foo i_final=3,j_final=6 1554281635112992012
```

Original input:

```text
counter,host=bar i=1,j=4 1554281633101153300
counter,host=foo i=1,j=4 1554281633099323601
counter,host=bar i=2,j=5 1554281634107980073
counter,host=foo i=2,j=5 1554281634105931116
counter,host=bar i=3,j=6 1554281635115090133
counter,host=foo i=3,j=6 1554281635112992012
```