Helm interface in cdk8s
Project description
Why
I still have two fundamental problems with using helm now that tiller is gone. The first is that it sometimes requires you to put secrets in config files. There's no good universal way to specify my manifests with shared secrets without having them show up in my repositories. The second is that there's no good way to manage global variables within a massive project. Sure it's possible with subcharts, but I've found it gets really messy really quickly.
I like using cdk8s for my own images and deployments. I think it's clean and easy way to follow best practices, and with the help of jedi and autocompletion it's super easy to write things that you know will work right.
Installing
pipenv install cdhelm
to add cdhelm to your Pipfile.
from cdhelm import Helm
in your cdk8s file.
Workflow
This library allows you to specify a helm chart within cdk8s. It's pretty flexible with how it lets you organize your project.
You can define a Helm object like so:
prometheus = Helm(
name="prometheus",
repo="https://kubernetes-charts.storage.googleapis.com",
chart="prometheus",
version="11.12.0",
namespace="monitoring,
path=".charts",
)
namespace
defaults to default
and path
defaults to .charts
. Otherwise
everything else is required. version
is explicitly required because I can't
think of a reason you should ever rely on latest
implicitly.
This returns a Helm object that you can interact with. Creating the object
pulls the chart to path
and populates object.values
with the default values
file. The is currently no option to not pull the whole chart.
Adding values
You can specify new values either through a values file like with normal helm charts.
prometheus.updateFile("configs/prometheus.yaml")
Or you can specify new values through a python dictionary corrisponding to the values file.
prometheus_conf = {
"alertmanager": {
"baseURL": "https://alert." + mydomain
},
"server": {
"baseURL": "https://prom." + mydomain
},
}
prometheus.update(prometheus_conf)
Both actions only update exactly what is specified so it's possible to use both together even on variables within the same scope.
Extracting values or working entirely within cdk8s
It is possible to create a cdk8s chart from the Helm object. This chart will bake in the current values. Please update your values before returning a chart.
monitoring_app = App(outdir=dist + "monitoring")
prometheus_chart = prometheus.Chart(monitoring_app, "prometheus")
You can use this to pass along information from the chart such as services and
ports to other parts of your code. This is a good way to interface different
parts of your server without hardcoding values. You may synth the app the chart
is a part of and work with helm as though it's just another part of your cdk8s
implementation. Every part of the chart will be rendered out using helm template
.
monitoring_app.synth()
Working outside of cdk8s
It is also possible to keep the benefits of having helm specified in your cdk8s
code while still letting helm do its thing. You can have cdk8s write out a
values.yaml
file so you can manage helm manually. render()
will overwrite
the default rendered values path.
prometheus.render("helm/myvalues.yaml")
Or you can have cdk8s render out a helper script to the specified folder to work with exactly the version and values you already specified.
prometheus.write_controls("_charts")
The rendered script should work with any helm command that works with a
specific chart. Use -h
with the chart to see a help printout.
All commands will be given the following options when applicable:
--namespace monitoring
--repo https://kubernetes-charts.storage.googleapis.com
--version 11.12.0
--values /home/myhome/myproject/.charts/prometheus-11.12.0/prometheus/rendered-values.yaml
[NAME] => prometheus
[CHART] => prometheus
[RELEASE] => prometheus
All other given options will be appended
Use -h as an appended option for any helm command
Available Helm commands:
help
history
install
rollback
show all
show chart
show readme
show values
status
template
test
uninstall
upgrade
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.
Source Distribution
Built Distribution
File details
Details for the file cdhelm-0.1.4.tar.gz
.
File metadata
- Download URL: cdhelm-0.1.4.tar.gz
- Upload date:
- Size: 5.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.24.0 setuptools/50.3.0 requests-toolbelt/0.9.1 tqdm/4.49.0 CPython/3.8.5
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 8d72d8180baa6baed1b3b888ba2d8500f55ed93031c7a23cbd9cb560aaf837bc |
|
MD5 | 704bf4eba843fb0ef3981259dda3a3d4 |
|
BLAKE2b-256 | a14e8da08acd1a51806b4295cfdeec23e81c974d43dff0e79e753d3dd573b2c2 |
Provenance
File details
Details for the file cdhelm-0.1.4-py3-none-any.whl
.
File metadata
- Download URL: cdhelm-0.1.4-py3-none-any.whl
- Upload date:
- Size: 8.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.24.0 setuptools/50.3.0 requests-toolbelt/0.9.1 tqdm/4.49.0 CPython/3.8.5
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 0e495feb20c34eb8b3f78068037eab261e2bc155402c8cf23fc9f62f5201749b |
|
MD5 | 1d9927b2b8e0cc61a5de78d4b3c9f655 |
|
BLAKE2b-256 | 3f4743f9fca98da5066580a20c9be9b71e31ec696fa13d4a9caa317ac1f29206 |