telegraf-pyplug 0.3.0

Telegraf_pyplug is a software library to simplify and standardize the development of python input plugins for the Telegraf

Problem

Telegaf is a plugin-driven agent for collecting, processing, aggregating, and writing metrics. Custom input plugins collect metrics from the system, services, or 3rd party APIs and outputs them in the InfluxDB line protocol format.

• Printing metrics in the InfluxDB line protocol format is a bit complicated, and it's easy to make mistakes.
• There is no standard way to develop Telegraf plugins in Python. Maintaining a lot of plugins designed in different ways becomes hell.

Solution

Telegraf_pyplug is a free and open-source software library to simplify and standardize the development of python input plugins for the Telegraf.

Usage

One field:

from telegraf_pyplug.main import print_influxdb_format


METRIC_NAME: str = 'jumping_sheep'
METRIC_COUNT: int = 321


def main() -> None:
    print_influxdb_format(measurement=METRIC_NAME, fields={'count': METRIC_COUNT})


if __name__ == '__main__':
    main()

Outputs:

jumping_sheep count=321

add_timestamp argument:

from telegraf_pyplug.main import print_influxdb_format


METRIC_NAME: str = 'jumping_sheep'
METRIC_COUNT: int = 321


def main() -> None:
    print_influxdb_format(measurement=METRIC_NAME, fields={'count': METRIC_COUNT}, add_timestamp=True)


if __name__ == '__main__':
    main()

Outputs:

jumping_sheep count=321 1599846911207090944

One field, One tag:

from telegraf_pyplug.main import print_influxdb_format


METRIC_NAME: str = 'jumping_sheep'
METRIC_COUNT: int = 321
METRIC_COLOR: str = 'white'


def main() -> None:
    print_influxdb_format(measurement=METRIC_NAME, tags={'color': METRIC_COLOR}, fields={'count': METRIC_COUNT})


if __name__ == '__main__':
    main()

Outputs:

jumping_sheep,color=white count=321

Multiple fields and tags, nano_timestamp argument:

from datetime import datetime
from typing import Dict

from telegraf_pyplug.main import print_influxdb_format, datetime_tzinfo_to_nano_unix_timestamp


METRIC_NAME: str = 'jumping_sheep'
METRIC_FIELDS: Dict[str, int] = {'count': 321, 'height_m': 1.5}
METRIC_TAGS: Dict[str, str] = {'color': 'white', 'name': 'sweater'}
METRIC_DATE: str = '01.01.2020 03:00:00+0300'


def main() -> None:
    date: datetime = datetime.strptime(METRIC_DATE, '%d.%m.%Y %H:%M:%S%z')

    print_influxdb_format(
        measurement=METRIC_NAME,
        tags=METRIC_TAGS,
        fields=METRIC_FIELDS,
        nano_timestamp=datetime_tzinfo_to_nano_unix_timestamp(date)
    )


if __name__ == '__main__':
    main()

Outputs:

jumping_sheep,color=white,name=sweater count=321,height_m=1.5 1577836800000000000

More advanced examples can be found in the examples_dir.

Installation

Telegraf_PyPlug can easily be installed with pip.

Mac/Linux

pip install --upgrade telegraf_pyplug

Windows

python -m pip install --upgrade telegraf_pyplug

License

Telegraf_PyPlug is under MIT license. See the LICENSE file for the full license text.