Custom Metrics

Note

This example is a replication of an Argo Workflow example in Hera. The upstream example can be found here.

HeraYAML

from hera.workflows import (
    Container,
    Counter,
    Gauge,
    Histogram,
    Label,
    Parameter,
    Steps,
    Workflow,
    models as m,
)

random_int = Container(
    name="random-int",
    image="alpine:latest",
    command=["sh", "-c"],
    args=["RAND_INT=$((1 + RANDOM % 10)); echo $RAND_INT; echo $RAND_INT > /tmp/rand_int.txt"],
    metrics=[
        Histogram(
            name="random_int_step_histogram",
            help="Value of the int emitted by random-int at step level",
            when="{{status}} == Succeeded",
            buckets=[2.01, 4.01, 6.01, 8.01, 10.01],
            value="{{outputs.parameters.rand-int-value}}",
        ),
        Gauge(
            name="duration_gauge",
            labels=Label(key="name", value="random-int"),
            help="Duration gauge by name",
            realtime=True,
            value="{{duration}}",
        ),
    ],
    outputs=Parameter(
        name="rand-int-value", global_name="rand-int-value", value_from=m.ValueFrom(path="/tmp/rand_int.txt")
    ),
)
flakey = Container(
    name="flakey",
    image="python:alpine3.6",
    command=["python", "-c"],
    args=["import random; import sys; exit_code = random.choice([0, 1, 1]); sys.exit(exit_code)"],
    metrics=Counter(
        name="result_counter",
        labels=[Label(key="name", value="flakey"), Label(key="status", value="{{status}}")],
        help="Count of step execution by result status",
        value="1",
    ),
)

with Workflow(
    generate_name="hello-world-",
    entrypoint="steps",
    metrics=Gauge(
        name="duration_gauge",
        labels=Label(key="name", value="workflow"),
        help="Duration gauge by name",
        realtime=True,
        value="{{workflow.duration}}",
    ),
) as w:
    with Steps(
        name="steps",
        metrics=Gauge(
            name="duration_gauge",
            labels=Label(key="name", value="steps"),
            help="Duration gauge by name",
            realtime=True,
            value="{{duration}}",
        ),
    ):
        random_int()
        flakey()

apiVersion: argoproj.io/v1alpha1
kind: Workflow
metadata:
  generateName: hello-world-
spec:
  entrypoint: steps
  metrics:
    prometheus:
    - gauge:
        realtime: true
        value: '{{workflow.duration}}'
      help: Duration gauge by name
      labels:
      - key: name
        value: workflow
      name: duration_gauge
  templates:
  - metrics:
      prometheus:
      - gauge:
          realtime: true
          value: '{{duration}}'
        help: Duration gauge by name
        labels:
        - key: name
          value: steps
        name: duration_gauge
    name: steps
    steps:
    - - name: random-int
        template: random-int
    - - name: flakey
        template: flakey
  - container:
      args:
      - RAND_INT=$((1 + RANDOM % 10)); echo $RAND_INT; echo $RAND_INT > /tmp/rand_int.txt
      command:
      - sh
      - -c
      image: alpine:latest
    metrics:
      prometheus:
      - help: Value of the int emitted by random-int at step level
        histogram:
          buckets:
          - 2.01
          - 4.01
          - 6.01
          - 8.01
          - 10.01
          value: '{{outputs.parameters.rand-int-value}}'
        name: random_int_step_histogram
        when: '{{status}} == Succeeded'
      - gauge:
          realtime: true
          value: '{{duration}}'
        help: Duration gauge by name
        labels:
        - key: name
          value: random-int
        name: duration_gauge
    name: random-int
    outputs:
      parameters:
      - globalName: rand-int-value
        name: rand-int-value
        valueFrom:
          path: /tmp/rand_int.txt
  - container:
      args:
      - import random; import sys; exit_code = random.choice([0, 1, 1]); sys.exit(exit_code)
      command:
      - python
      - -c
      image: python:alpine3.6
    metrics:
      prometheus:
      - counter:
          value: '1'
        help: Count of step execution by result status
        labels:
        - key: name
          value: flakey
        - key: status
          value: '{{status}}'
        name: result_counter
    name: flakey

Comments