cmon collects metrics on running containers and saves them to an HDF5
file.
Clone this repository and run poetry install. Afterwards you will be able to
invoke cmon as poetry run cmon.
You may alternatively build a Python wheel for cmon and install it system
wide as follows, assuming you are in the repository root:
poetry build
pip install dist/cmon-0.1.0-py3-none-any.whlFrom then on you should have cmon available in your path.
python -m cmon -h
usage: cmon monitor [-h] [-c COMPOSE] [-p PERIOD] [-o OUTPUT]
[--buffer BUFFER]
options:
-h, --help show this help message and exit
-c COMPOSE, --compose COMPOSE
If provided, filter by Docker Compose project
-p PERIOD, --period PERIOD
Sampling period [default=0.1s]
-o OUTPUT, --output OUTPUT
Output path [defaults to ./<timestamp>.hdf5]
--buffer BUFFER Length of measurement buffer [default=1024]
Experimental plotting based on Dash.
usage: cmon plot [-h] input_file
positional arguments:
input_file Path to HDF5 input file
options:
-h, --help show this help message and exit
The (interactive) plots should show up in http://localhost:8050.