I've noticed that we can crate a setter contains list of string based on kpt [documentation][1]. Then I found out that complex setter contains list of object is not supported based on [this github issue][1]. Since the issue itself mentioned that this should be supported in kpt function can we use it with the current kpt function version?
[1]: Kpt Apply Setters. https://catalog.kpt.dev/apply-setters/v0.1/
[1]: Setters for list of objects. https://github.com/GoogleContainerTools/kpt/issues/1533
I've discussed a bit with my coworkers, and turned out this is possible by doing the following setup:
apiVersion: apps/v1
kind: Deployment
metadata:
name: my-nginx
spec:
replicas: 4 # kpt-set: ${nginx-replicas}
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: "nginx:1.16.1" # kpt-set: nginx:${tag}
ports:
- protocol: TCP
containerPort: 80
---
apiVersion: v1
kind: MyKind
metadata:
name: foo
environments: # kpt-set: ${env}
- dev
- stage
---
apiVersion: v1
kind: MyKind
metadata:
name: bar
environments: # kpt-set: ${nested-env}
- key: some-key
value: some-value
After that we can define the following setters:
apiVersion: v1
kind: ConfigMap
metadata:
name: setters
data:
env: |-
- prod
- dev
nested-env: |-
- key: some-other-key
value: some-other-value
nginx-replicas: "3"
tag: 1.16.2
And then we can call the following command:
$ kpt fn render apply-setters-simple
I've send a Pull Request to the repository to add documentation about this.
My requirement is inside pod there is a file
location : /mnt/secrets-store/environment
In kubernetes manifest file i would like to set environment variable . Values contains above location flat file
pls share your thought how to achieve that
I have tried below option in the k8s yml file but not working
apiVersion: apps/v1
kind: Deployment
metadata:
name: sample-api
namespace: dev
spec:
replicas: 3
selector:
matchLabels:
app: sample-api
template:
metadata:
labels:
app: sample-api
aadpodidbinding: azure-pod-identity-binding-selector
spec:
containers:
- name: sample-api
image: sample.azurecr.io/sample:11129
imagePullPolicy: Always
ports:
- containerPort: 80
env:
- name: "ASPNETCORE_ENVIRONMENT"
value: "Kubernetes"
- name: NRIA_DISPLAY_NAME
value: $("/usr/bin/cat" "/mnt/secrets-store/environment")
apiVersion: v1
kind: Pod
metadata:
name: myapp-pod
labels:
app: myapp
spec:
containers:
- name: myapp-container
image: mysql:8.0
env:
- name: MYSQL_ROOT_PASSWORD
value: root
You can use a secret for storing that value and use that secret as env in the pod.
https://kubernetes.io/docs/tasks/inject-data-application/distribute-credentials-secure/#define-container-environment-variables-using-secret-data
I'm applying a Kubernetes CronJob.
So far it works.
Now I want to add the environment variables. (env: -name... see below)
While tryng to apply I get the error
unknown field "configMapRef" in io.k8s.api.core.v1.EnvVarSource
I don't like to set all singles variables here. I prefer to link the configmap to not to double the variables. How is it possible set a link to the configmap.yaml variables in a CronJob file, how to code it?
Frank
apiVersion: batch/v1beta1
kind: CronJob
metadata:
name: ad-sync
creationTimestamp: 2019-02-15T09:10:20Z
namespace: default
selfLink: /apis/batch/v1beta1/namespaces/default/cronjobs/ad-sync
spec:
concurrencyPolicy: Allow
failedJobsHistoryLimit: 1
successfulJobsHistoryLimit: 3
suspend: false
schedule: "0 */1 * * *"
jobTemplate:
metadata:
labels:
job: ad-sync
spec:
template:
spec:
containers:
- name: ad-sync
image: foo.azurecr.io/foobar/ad-sync
command: ["dotnet", "AdSyncService.dll"]
args: []
env:
- name: AdSyncService
valueFrom:
configMapRef:
name: ad-sync-service-configmap
restartPolicy: OnFailure
There is no such field configMapRef in env field instead there is a field called configMapKeyRef
in order to get more detail about kubernetes objects, its convenient to use kubectl explain --help
for example if you would like to check all of the keys and their types you can use following command
kubectl explain cronJob --recursive
kubectl explain cronjob.spec.jobTemplate.spec.template.spec.containers.env.valueFrom.configMapKeyRef
You should use configMapKeyRef for single value or configMapRef with envFrom
It works this way:
apiVersion: batch/v1beta1
kind: CronJob
metadata:
...
spec:
...
jobTemplate:
metadata:
...
spec:
template:
spec:
containers:
- name: ad-sync
...
envFrom:
- configMapRef:
name: ad-sync-service-configmap
command: ["dotnet", "AdSyncService.dll"]
There are two approaches, using valueFrom for individual values or envFrom for multiple values.
valueFrom is used inside the env attribute, like this:
spec:
template:
spec:
containers:
- name: ad-sync
image: foo.azurecr.io/foobar/ad-sync
command: ["dotnet", "AdSyncService.dll"]
args: []
env:
- name: AdSyncService
valueFrom:
configMapKeyRef:
name: ad-sync-service-configmap
key: log_level
envFrom is used direct inside the container attribure like this:
spec:
template:
spec:
containers:
- name: ad-sync
image: foo.azurecr.io/foobar/ad-sync
command: ["dotnet", "AdSyncService.dll"]
envFrom:
- configMapRef:
name: ad-sync-service-configmap
ConfigMap for reference:
apiVersion: v1
kind: ConfigMap
metadata:
name: ad-sync-service-configmap
namespace: default
data:
log_level: INFO
The main difference on both is:
valueFrom will inject the value of a a key from the referenced configMap
envFrom will inject All configMap keys as environment variables
The main issue with you example is that you used the configMapRef from envFrom inside the valueFrom where should actually be configMapKeyRef.
Also, the configMapKeyRef need a key attribute to identify where the data is comming from.
For more details, please check in this docs.
I can't seem to find an example of the correct syntax for inserting a environment variable along with the service name:
So I have a service defined as:
apiVersion: v1
kind: Service
metadata:
name: test
spec:
type: NodePort
ports:
- name: http
port: 3000
targetPort: 3000
selector:
app: test
I then use a secrets file with the following:
apiVersion: v1
kind: Secret
metadata:
name: test
labels:
app: test
data:
password: fxxxxxxxxxxxxxxx787xx==
And just to confirm I'm using envFrom to set that password as an env variable:
apiVersion: v1
kind: Deployment
metadata:
name: test
spec:
replicas: 1
template:
metadata:
labels:
app: test
spec:
containers:
- name: test
image: xxxxxxxxxxx
imagePullPolicy: Always
envFrom:
- configMapRef:
name: test
- secretRef:
name: test
ports:
- containerPort: 3000
Now in my config file I want to refer to that password as well as the service name itself - is this the correct way to do so:
apiVersion: v1
kind: ConfigMap
metadata:
name: test
labels:
app: test
data:
WORKING_URI: "http://somedomain:${password}#test"
The yaml configuration does not work the way you provided as an example.
If you want to setup Kubernetes with a complex configuration and
use variables or dynamic assignment to some of them, you have to
use an external parser to replace variable place holders. I use
bash and sed to accomplish it. I changed your config a bit:
apiVersion: v1
kind: ConfigMap
metadata:
name: test
labels:
app: test
data:
WORKING_URI: "http://somedomain:VAR_PASSWORD#test"
After saving, I created a simple shell script containing desired values.
#!/bin/sh
export PASSWORD="verysecretpwd"
cat deploy.yaml | sed "s/VAR_PASSWORD/$PASSWORD/g" | kubectl -f -