Draw diagrams from tf state files
Project description
drawtf
Draw diagrams which include Cloud resources using TF state files or without them. Inspired by the Diagrams package and a burning desire to not have to manually keep architecture diagrams updated, this was born.
Prerequisites
- Install Python
- Install Graphviz
Install
foo@bar:~$ python -m pip install --upgrade pip
foo@bar:~$ pip install drawtf
foo@bar:~$ drawtf --help
Usage: drawtf [OPTIONS]
Console script for drawtf.
Options:
--name TEXT The diagram name.
--state TEXT The tfstate file to run against.
--platform TEXT The platform to use 'azure' or 'aws', only
'azure' currently supported
--output-path TEXT Output path if to debug generated json
populated.
--json-config-path TEXT Config file path if populated.
--json-config-override-path TEXT
Config file with overrides path to merge
with main one if populated.
--verbose Add verbose logs.
--help Show this message and exit.
Usage
There are a few ways we can create diagrams here, all options on the CLI are optional, and it is basically just the order which you create them that draws a diagram.
Sample config File (app.json)
If we use a config file with the fields below, this will set the name for the designs title, import a state file, and add some custom components not in the state file, The final section at the bottom draws the lines between the resources.
{
"name": "Aggreko Application (All Resources)",
"state": "../../application.tfstate",
"components": [
{
"name": "Aggreko",
"type": "draw_custom",
"resource_group_name": "",
"attributes":{},
"components": [
{
"name": "InternalDB",
"type": "draw_custom",
"resource_group_name": "",
"custom": "diagrams.azure.database.DatabaseForMysqlServers",
"attributes": {}
},
{
"name": "App",
"type": "draw_custom",
"resource_group_name": "",
"custom": "diagrams.onprem.compute.Server",
"attributes": {}
},
{
"name": "AnotherApp",
"type": "draw_custom",
"resource_group_name": "",
"custom": "diagrams.onprem.compute.Server",
"attributes": {}
}
]
}
],
"links": [
{
"from": "App-draw_custom",
"to": "InternalDB-draw_custom",
"color": "darkgreen",
"label": "Write",
"type": "dotted"
},
{
"from": "AnotherApp-draw_custom",
"to": "InternalDB-draw_custom",
"color": "darkgreen",
"label": "Write",
"type": "dotted"
}
],
"excludes": []
}
foo@bar:~$ drawtf --json-config-path ./test/app.json
By running the command above pointing to the config file, this will set the name and grab other resources from the state file linked. Outputs from will create the design in the same sub-folder with the name app.png.
Override config File (app-subset.json)
Providing an override config alongside our main config file with the fields below, this will override the initial the designs title but still use the same state file and components from the original and attempt to join the links if all of the resources are available. You will notice an excludes section, if the keys for each resource added are in this list, then it will exclude those items.
{
"name": "Aggreko Application (Subset)",
"excludes": [
"AnotherApp-draw_custom"
]
}
foo@bar:~$ drawtf --json-config-path ./test/app.json --json-config-override-path ./test/app-subset.json
By running the command above pointing to the config file and override files, this will set the name from the override and grab other resources from the state file linked. Outputs from will create the design in the same sub-folder with the name app-subset.png.
Override config File and CLI overrides
foo@bar:~$ drawtf --json-config-path ./test/app.json --json-config-override-path ./test/app-subset.json --name "Aggreko Application (Sample)" --state ./test/app.tfstate --output-path ./test/sample --verbose
The command above, though using the same config files, can override all for the name, state file path and output path. Outputs from will create the design in the directory test with the name sample.png.
Github Actions Steps
Yes you can run this via GitHub actions or devops pipelines.
- name: Setup Graphviz
uses: ts-graphviz/setup-graphviz@v1
- name: Set up Python 3.7
uses: actions/setup-python@v4
with:
python-version: '3.7'
- name: Generate Diagram
run: |
python -m pip install --upgrade pip
pip install drawtf
drawtf --help
drawtf --json-config-path "./static/images/test/test.json"
Early days
Just an FYI, its early days here and is still a development style project. That said we are using for all of our projects internally using TF, but loads of resources types are still to be added.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.