Hey @Anibal Ulisses Can you try to follow the Kubernetes documentation for creating a .dockerconfigjson secret as outlined here to achieve the above:
https://kubernetes.io/docs/tasks/configure-pod-container/pull-image-private-registry/
Hey, @Anibal Ulisses Sorry for the above. I misunderstood the question. The above article I sent is for a different use case.
You will need to do the below:
1) On master and worker nodes create $URL directory at /etc/docker/certs.d/ where $URL is replaced with their registry hostname and port:
sudo mkdir /etc/docker/certs.d/example.com:5000
Copy (scp or download) registry.crt to /etc/docker/certs.d/example.com:5000/registry.crt
Hi @AnishWalia20
Great, I need to restart the cluster or they identify the new certificates and restart automatic?
Hey @Anibal Ulisses . Sorry I couldn’t reach out. I am not sure about the restart thing.
Did you try the above? . Let me know if you need help with anything else.
Hi @AnishWalia20
Problem solved…
On my notebook docker installation I already configured the certificate like you described:
1) On master and worker nodes create $URL directory at /etc/docker/certs.d/ where $URLis replaced with their registry hostname and port:
sudo mkdir /etc/docker/certs.d/example.com:5000
Copy (scp or download) registry.crt to /etc/docker/certs.d/example.com:5000/registry.crt
But I found two problem that after I understand I have success to configure the certificate chain:
- Need to run this command to reload the new settings:
systemctl daemon-reload && systemctl restart docker
- Need to install the certificate on all Master and Worker nodes, and run step 1 on each one.
Another point that I faced, unfortunately I have a proxy between my server and internet. When the karbon are deployed they didn’t get the “whitelist” configure at Prism Element to include on the NO_PROXY variable, they only get the PROXY_HTTP(S) information. I need to adjust on all nodes.
No my certificate chain are running fine :-)
Thank you.
Anibal
Ahh, that is amazing @Anibal Ulisses . Glad that it worked out. And also thanks for the side note about restarting docker daemon on all the master and worker nodes to make the settings persistent.