Grafana 的网格工具包

根据数百个环境的多种变体代码生成相似但不同的仪表板。基于 grafanalib

安装和快速入门

❯❯❯ virtualenv .py3
❯❯❯ source .py3/bin/activate

❯❯❯ pip install noroutine-grit

❯❯❯ python -m grit -h

# Inspect examples
❯❯❯ python -m grit inspect --module examples.dashboards
# Example output
---
name: examples.dashboards
folders:
- jmeter
- mongodb
variations:
  environment:
  - dev
  - prod
  - qa
  - test

# Generate dashboards for all environments
❯❯❯ python -m grit generate --module examples.dashboards --out 'out/{environment}' --var 'environment=*' 
# Example output 
Generating out/dev
Generating out/prod
Generating out/qa
Generating out/test

概述

示例仪表板

GritDash(
    uid="human-id",
    version=8,
    title="your dashboard title",
    description="your dashboards description",
    tags=[
        'tag1',
        'tag2'
    ],
    timezone="browser",
    # Apply datasource to all panels in the dashboard
    dataSource="Prometheus",
    stack=Stack(
        row7(panel1),
        # Use simple utilities, row6 is row of panels of height 6
        row6(panel1, panel2),
        # Autosize panels, this will be 3 equal panels in the row
        row5(panel1, panel2, panel3),
        # ... or 4, if you want, no need to calculate coordinates
        row4(panel1, panel2, panel3, panel4),
        # ... sometimes less is more
        row3(panel1, panel2, panel3, panel4, panel5),
    )
)

变化

变体允许从相同的代码库创建不同的仪表板集合，例如每个环境的变体可以通过环境变体来实现。

您可以创建多个变体，并快速生成数百个仪表板

# Variation is just subclass of Variation
class Environment(Variation):
  color: str = "green"

Environment(name="dev", color="blue")
Environment(name="qa")
Environment(name="test")
Environment(name="prod", color="red")

编写仪表板时，您可以通过调用访问特定的变体值<VariationClass>.resolve()

# Illustrative code
environment = Environment.resolve()
print(environment.color)
print(environment.name)

您可以根据需要创建尽可能多的变体并将它们组合起来

class Environment(Variation):
  color: str = "green"

Environment(name="dev", color="blue")
Environment(name="qa")
Environment(name="test")
Environment(name="prod", color="red")

class Turn(Variation):
  pass

class Taste(Variation):
  pass

class Animal(Variation):
  pass

Turn(name="first")
Turn(name="second")
Turn(name="third")

Taste(name="sweet")
Taste(name="sour")

Animal(name="rabbit")
Animal(name="donkey")
Animal(name="turtle")
Animal(name="rat")

并在生成过程中使用它们，您可以根据自己的喜好模板输出目录结构！

❯❯❯ python -m grit generate --module grafana.dashboards --out 'out/{turn}-{taste}-{animal}-company/{environment}' --var environment=qa environment=prod animal=* turn=first taste=sweet
Generating out/first-sweet-donkey-company/qa
Generating out/first-sweet-donkey-company/prod
Generating out/first-sweet-rabbit-company/qa
Generating out/first-sweet-rabbit-company/prod
Generating out/first-sweet-rat-company/qa
Generating out/first-sweet-rat-company/prod
Generating out/first-sweet-turtle-company/qa
Generating out/first-sweet-turtle-company/prod

命令行

检查仪表板

❯❯❯ python -m grit inspect --module dashboards
---
name: dashboards
folders:
- jmeter
- mongodb
variations:
  environment:
  - dev
  - nab-dev
  - nab-prod
  - prod
  - qa
  - test

发布到 Grafana

# Needs .env file with some vars
❯❯❯ python -m grit publish --module grafana.dashboards --var environment=dev

生成到输出目录

❯❯❯ python -m grit generate --module grafana.dashboards --out 'out/{environment}' --var environment=*
Generating out/dev
Generating out/nab-dev
Generating out/nab-prod
Generating out/prod
Generating out/qa
Generating out/test

实用程序

行

你有一些方便的实用程序

预定义的行高

各自高度的每个创建行

• row3
• row4
• row5
• row6
• row7
• row8

如何发展

virtualenv .py3
source .py3/bin/activate
pip install -r requirements.txt
export PYTHONPATH=src
python -m grit --version