Version 7.1 CDK Deployment on Docker Desktop

[Thomas Liebeck]

Maybe someone of you made it to deploy the CDK 7.1 to Docker Desktop Mac BigSur Intel processor.
For me “./cdk install -n mynamespace -f dev.example.com” fails to deploy the ds-idrepo-0 stateful set

Here the output
—————
Checking secret-agent operator and related CRDs: secret-agent CRD found in cluster.
Checking ds-operator and related CRDs: ds-operator CRD found in cluster.

Installing component(s): [‘quickstart’]

Deploying base.yaml. This is a one time activity.
configmap/dev-utils unchanged
configmap/platform-config unchanged
Warning: networking.k8s.io/v1beta1 Ingress is deprecated in v1.19+, unavailable in v1.22+; use networking.k8s.io/v1 Ingress
ingress.networking.k8s.io/end-user-ui configured
ingress.networking.k8s.io/forgerock configured
ingress.networking.k8s.io/ig-web configured
ingress.networking.k8s.io/login-ui configured
ingress.networking.k8s.io/platform-ui configured
secretagentconfiguration.secret-agent.secrets.forgerock.io/forgerock-sac configured

Deploying ds.yaml. This is includes all directory resources.
directoryservice.directory.forgerock.io/ds-idrepo configured

Waiting for DS deployment. This can take a few mins. First installation takes longer.
Waiting for 1 pods to be ready…
error: timed out waiting for the condition
Traceback (most recent call last):
File “/Users/tliebe/git/forgeops/bin/./cdk”, line 268, in <module>
main()
File “/Users/tliebe/git/forgeops/bin/./cdk”, line 255, in main
deploy_quickstart(namespace, fqdn, None)
File “/Users/tliebe/git/forgeops/bin/./cdk”, line 106, in deploy_quickstart
utils.wait_for_ds(ns, ‘ds-idrepo’)
File “/Users/tliebe/git/forgeops/bin/utils.py”, line 93, in wait_for_ds
run(‘kubectl’, f’-n {ns} rollout status –watch statefulset {directoryservices_name} –timeout=300s’)
File “/Users/tliebe/git/forgeops/bin/utils.py”, line 38, in run
_r = subprocess.run(runcmd.split(), stdout=stdo_pipe, stderr=stde_pipe,
File “/usr/local/Cellar/[email protected]/3.9.6/Frameworks/Python.framework/Versions/3.9/lib/python3.9/subprocess.py”, line 528, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command ‘[‘kubectl’, ‘-n’, ‘tliebe’, ‘rollout’, ‘status’, ‘–watch’, ‘statefulset’, ‘ds-idrepo’, ‘–timeout=300s’]’ returned non-zero exit status 1.

Any ideas?

[Thomas Liebeck]

Found it after some kubernetes reading.
Docker-Desktop only provides the storage class “hostpath”
kubectl get storageclass
NAME PROVISIONER RECLAIMPOLICY VOLUMEBINDINGMODE ALLOWVOLUMEEXPANSION AGE
hostpath (default) docker.io/hostpath Delete Immediate false 5d22h

The configured storage class for PVC was “standard”
To fix this I replaced the storage class “standard” with “hostpath” in each yaml file.
To identify the files I executed from forgeops/bin directory “grep -r “storageClassName: standard” ./*”

Than I performed a cdk delete –namespace my-namespace and a cdk install –namespace my-namespace -f dev.example.com

Now everything is up and running. If it works I don’t know because this was just to overcome the deployment hurdle on docker-desktop.

[Author]

Posts