Welcome to the Linux Foundation Forum!

Validation of cluster configuration

prenticect
prenticect Posts: 21
in LFS258 Class Forum

My system is not as expected according to the lab, so I now have two parts and I am checking to see if I am off by a little or a lot.

(All on the Master server)

1 -

student@lfs458-node-1a0a:~$ kubectl get node
NAME STATUS ROLES AGE VERSION
lfs458-node-1a0a Ready master 28m v1.14.1
lfs458-worker Ready 50s v1.14.1

I only have the master node, no worker

2 -

student@lfs458-node-1a0a:~$ kubectl get pods --all-namespaces
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system calico-etcd-jlgwr 1/1 Running 0 6m
kube-system calico-kube-controllers-74b888b647-wlqf5 1/1 Running 0 6m
kube-system calico-node-tpvnr 2/2 Running 0 6m
kube-system coredns-78fcdf6894-nc5cn 1/1 Running 0 17m
kube-system coredns-78fcdf6894-xs96m 1/1 Running 0 17m

I have many in the COntainerCreating stage
student@ubuntu:~$ kubectl get pods --all-namespaces
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-system calico-node-wgfmw 0/2 ContainerCreating 4 2d1h
kube-system coredns-fb8b8dccf-7dvgf 0/1 ContainerCreating 322 2d1h
kube-system coredns-fb8b8dccf-psbj2 0/1 ContainerCreating 318 2d1h
kube-system etcd-ubuntu 0/1 Pending 0 0s
kube-system kube-proxy-ncm5q 0/1 ContainerCreating 1 2d1h

Now I think something is wrong. The kubeadm join command worked, I generated a key and all those steps went as stated. So either I missed something or I did something wrong. Any suggestions? Or is this all fine and I need to delete the two coredns pods and let them get re-created?

Thank you.

Comments

  • chrispokorni
    chrispokorni Posts: 2,605
    edited June 2019

    Hi,
    The ContainerCreating state indicates that the images cannot be pulled from the Docker repository, and the fact that you can't see your worker node indicates that the join did not work.
    Overall this is a networking issue with your VMs/nodes. In order for Kubernetes to properly work, the infrastructure networking has to be setup correctly. Kubernetes expects that the nodes can see each other and communicate with each other, and also each node should have access to the internet. These are settings you need to revise in VMware for each VM/node.
    There may be a firewall enabled at the guest OS level in Ubuntu, which may again block some traffic. I recommend verifying that also.
    Regards,
    -Chris

  • prenticect
    prenticect Posts: 21

    Fair enough Chris. Thank you. I will work on that and retry the join. Thanks for the clarification on the docker part too.

  • prenticect
    prenticect Posts: 21

    One the worker node I got this message. I coded in the actual IP address so I do not know why this message is referring to localhost

    student@ubuntu:~$ kubectl get nodes
    The connection to the server localhost:8080 was refused - did you specify the right host or port?

  • coop
    coop Posts: 917

    Hi:

    Just to explain this is not a real time tech support post. New students sometimes don't fully grok that.

    You have 20 posts in the last few days, and the moderators have been rather gracious in spending the time to explain things. While this is their job, perhaps you could hold off on posting until you study the documentation more carefully and try various things, as well as search through the archives of the forum.

    Of course, if you see things that are wrong in the course material, do not hesitate to promptly post about that as it gladly accepted.

    Thanks for your consideration.

  • prenticect
    prenticect Posts: 21

    Thanks coop.

    Yes, I assumed given a forum the responses would not be real-time. I have been using the search feature before posting as I hope someone has already asked the question I have and the answer is already there. And I do admit one of the issues was I misread the prompt (a product of doing this effort after working hours and getting tired)

    I also now have a teammate working on the course material starting soon. He is doing the cloud version whereas I am doing the VMWare version. Hopefully as he goes through this we can work together and solve our own issues.

    For reading documentation, I have gone through the class material up to where is says do lab 3, up to lab 3.5. All the labs are self documenting. Thus it is difficult when something seemly does not work to know what is wrong or not wrong or missing information or even assumed aspects. {Within many kubernetes training programs I have been trying to get through there are many assumptions around environments, packages, knowledge of what just should happen, differences in Linix, kubernetes products. I understand and realize this environment is changing quickly and continuously which is also why I do appreciate pointing to other points of reference.}

    If I have over-stepped the usage of this forum, I do apologize and again thank the moderators for responding.

  • chrispokorni
    chrispokorni Posts: 2,605

    Hi @prenticect,
    Please review the exercise to make sure that you ran that command on the correct node. Running it on the wrong node produces the "... connection ... refused ..." error output.
    Renaming your nodes may also help in the future. Rather than having both named "ubuntu", one could be named "master" and the other "worker". When working on both nodes at the same time this will save you some headaches.
    Regards,
    -Chris

Categories

Upcoming Training