DEV Community

Cover image for Prometheus: Python metrics (with Docker and Gitlab CI)
cosckoya
cosckoya

Posted on

Prometheus: Python metrics (with Docker and Gitlab CI)

Prometheus is a metric based monitoring platform also it's one of all-time my favorite tools ever. And from a time ago I was thinking to develop some project with it. What's my plan? I want to create my multi-language application cluster with Prometheus monitoring, and then add some Grafana Loki, Cortex and Thanos integrations.

What's the first step?. Integrate some Prometheus metrics library with a basic python app. To reach this, I just use the Prometheus "client_python" sample, and make a container with it and push into a public registry. So...

First step! Create a Gitlab repository

Second step! Create an "app" folder and copy-paste this code into a "main.py":

from prometheus_client import start_http_server, Summary
import random
import time

# Create a metric to track time spent and requests made.
REQUEST_TIME = Summary('request_processing_seconds', 'Time spent processing request')

# Decorate function with metric.
@REQUEST_TIME.time()
def process_request(t):
    """A dummy function that takes some time."""
    time.sleep(t)

if __name__ == '__main__':
    # Start up the server to expose the metrics.
    start_http_server(8000)
    # Generate some requests.
    while True:
        process_request(random.random())
Enter fullscreen mode Exit fullscreen mode

Add a "requisites.txt" file with this content:

prometheus_client
Enter fullscreen mode Exit fullscreen mode

Add a "Dockerfile" like this one:

FROM python:3.9-alpine

WORKDIR /app
COPY . .
RUN pip install -r requisites.txt

RUN chmod u+x main.py

ENTRYPOINT ["/app/main.py"]
Enter fullscreen mode Exit fullscreen mode

Third Step! Create a Gitlab CI with Container Registry pipeline. To manage this task, I created a ".gitlab-ci.yaml" file in the repository basepath:

stages:
- build

image: docker:stable

services:
- docker:dind

build:
  stage: build
  when: on_success
  only:
  - master
  script:
  - docker login -u $CI_REGISTRY_USER -p $CI_JOB_TOKEN $CI_REGISTRY
  - docker build -f app/Dockerfile -t $CI_REGISTRY_IMAGE app
  - docker push $CI_REGISTRY_IMAGE
Enter fullscreen mode Exit fullscreen mode

This repository should look like this:

├── app
│   ├── Dockerfile
│   ├── main.py
│   └── requisites.txt
├── .gitlab-ci.yml
└── README.md
Enter fullscreen mode Exit fullscreen mode

Now, let commit all the files and wait until the pipeline is finished.

Four and last step! Run that image and scrape some metrics:

Run docker container as a daemon, with "python-prom" name, listening on 8000/TCP and delete that container when the image stops:

$> docker run -d --rm --name python-prom -p 8000:8000 registry.gitlab.com/cosckoya/python-prom
Enter fullscreen mode Exit fullscreen mode

Check that the image is up & running (also listening on that port)

$> docker ps

CONTAINER ID   IMAGE                                      COMMAND          CREATED              STATUS              PORTS                                       NAMES
380cc1a00a8b   registry.gitlab.com/cosckoya/python-prom   "/app/main.py"   About a minute ago   Up About a minute   0.0.0.0:8000->8000/tcp, :::8000->8000/tcp   python-prom
Enter fullscreen mode Exit fullscreen mode

And scrape those metrics!

$> curl localhost:8000

# HELP python_gc_objects_collected_total Objects collected during gc
# TYPE python_gc_objects_collected_total counter
python_gc_objects_collected_total{generation="0"} 309.0
python_gc_objects_collected_total{generation="1"} 43.0
python_gc_objects_collected_total{generation="2"} 0.0
# HELP python_gc_objects_uncollectable_total Uncollectable object found during GC
# TYPE python_gc_objects_uncollectable_total counter
python_gc_objects_uncollectable_total{generation="0"} 0.0
python_gc_objects_uncollectable_total{generation="1"} 0.0
python_gc_objects_uncollectable_total{generation="2"} 0.0
# HELP python_gc_collections_total Number of times this generation was collected
# TYPE python_gc_collections_total counter
python_gc_collections_total{generation="0"} 36.0
python_gc_collections_total{generation="1"} 3.0
python_gc_collections_total{generation="2"} 0.0
[..]
Enter fullscreen mode Exit fullscreen mode

Our Python Prometheus base image is done.

On my next post, I will create a Go or Java base prometheus application and all three of these base images will be deployed into a Kubernetes cluster alongside a Prometheus and create some Prometheus workbench with them.

Top comments (0)