I installed a Kubernetes cluster by following the instruction here:
https://github.com/GoogleCloudPlatform/kubernetes/blob/master/docs/getting-started-guides/vagrant.md
Everything looks fine the first time. I'm able to see the nodes, pods, deploy new pods, etc.
The problem shows up when I stop the cluster and try to start it again. I'm restarting the cluster as indicated on the documentation:
vagrant halt
./cluster/kube-up.sh
When I do that I see the following error:
Comment: Source file salt://kubelet/kubeconfig not found
...
Minion did not return. [No response]
Then, when I check the status of nodes it says the minion is NotReady.
If I have VirtualBox open while I run kube-up.sh, I see that the error is thrown before the minion VM is started. So it sounds like the minion is not running when it tries to configure it. That's just an observation, not sure what's the problem.
In order to solve this issue I have to destroy the cluster and create it again, what downloads and install everything again, making it very slow to use.
I found this problem on GitHub:
https://github.com/GoogleCloudPlatform/kubernetes/issues/9270
Here it was suggested to use the code in HEAD. I did that and now it is working fine.
Related
We have our k8s cluster set up with our app, including a neo4j DB deployment and other artifacts. Overnight, we've started facing an issue in our GKE cluster when trying to enter or interact somehow with any pod running in the cluster. The following screenshot shows a sample of the error we get.
issued command
error: unable to upgrade connection: Authorization error (user=kube-apiserver, verb=create, resource=nodes, subresource=proxy)
Our GKE cluster is created as standard (no autopilot) and the versions are
Node pool details
cluster basics
As said before it was working fine regardless of the warning about the versions. However, we haven't been able yet to identify what could have changed between the last time it worked, and now.
Any clue on what authorization setup might have been changed making it incompatible now is very welcomed
I have been setting up multi node kubernetes cluster using kubeadm.Setup included 1 master and worker node each. I have created the VM using vagrant.
I followed the docs,
https://kubernetes.io/docs/setup/independent/install-kubeadm/
https://kubernetes.io/docs/setup/independent/create-cluster-kubeadm
Created 2 VM's using vagrant
IP: Master- 192.168.33.10 , Worker- 192.168.1.21 (Both host only network)
I have experienced 2 scenarios,
Case 1:
Ran kubeadm init --pod-network-cidr=10.244.0.0/16 successfully with all pods running.
Installed "Canal" pod network add on.
Followed all the instructions given at the end of the successfull kubeadm init command.
SSH into 2nd VM and ran kubeadm join .. command and I am struck at "[preflight] Running pre-flight checks"
Case 2:
Did the same process with tag --apiserver-advertise-address=192.168.33.10
Successfully ran the command kubeadm init --apiserver-advertise-address=192.168.33.10
But when I ran the command kubectl get nodes it only showed the master node. (expected the worker node to show too).
Kindly help me understand how can I complete this setup. Thank you.
I have github repository which does exactly what you want. I am pretty sure that you will get idea from it. If anything is not clear, please update with comment or original post.
I'm trying to give fabric8 a shot but I'm having issues getting it to start on a local machine running minikube and virtualbox (I've attempted this on Linux and OSX. I'm able to get all but one of the pods to start (after manually increasing minikube's VM ram to 8GB). The expose controller won't start and is giving me the following error in the logs:
I0415 14:29:43.431944 1 exposecontroller.go:47] Using build: '2.3.2'
F0415 14:29:43.492059 1 exposecontroller.go:66] failed to create new strategy: failed to create node port expose strategy: failed to list nodes: nodes is forbidden: User "system:serviceaccount:fabric8:exposecontroller" cannot list nodes at the cluster scope
Here are the commands I'm running:
minikube start --cpus=5 --disk-size=50g --memory=8000
curl -sS http://get.fabric8.io/download.txt | bash
gofabric8 start
I also tried creating OAuth secret via GitHub (using bogus IP address info for the redirect URL) but this doesn't make sense to me because I don't have a domain... Then I ran these:
minikube start --vm-driver=xhyve --cpus=5 --disk-size=50g --memory=8000
minikube addons enable ingress
gofabric8 deploy --package system -n fabric8
That resulted in the exposecontroller working but then additional pods (keycloak, for example) were created but failed to start.
I've spent hours trying to get this to work and am about to give up. The documentation on GitHub differs from fabric8's site documentation and I just can't get it to work. If someone is able to help, I would greatly appreciate it.
Note:
I've attempted to follow the instructions here:
http://fabric8.io/guide/getStarted/gofabric8.html
Additionally, I attempted to follow this:
https://github.com/fabric8io/fabric8-platform/blob/master/INSTALL.md
I have an weave network plugin.
inside my folder /etc/cni/net.d there is a 10-weave.conf
{
"name": "weave",
"type": "weave-net",
"hairpinMode": true
}
My weave pods are running and the dns pod is also running
But when i want to run a pod like a simple nginx wich will pull an nginx image
The pod stuck at container creating , describe pod gives me the error , failed create pod sandbox.
When i run journalctl -u kubelet i get this error
cni.go:171] Unable to update cni config: No networks found in /etc/cni/net.d
is my network plugin not good configured ?
i used this command to configure my weave network
kubectl apply -f https://git.io/weave-kube-1.6
After this won't work i also tried this command
kubectl apply -f “https://cloud.weave.works/k8s/net?k8s-version=$(kubectl version | base64 | tr -d ‘\n’)”
I even tried flannel and that gives me the same error.
The system i am setting kubernetes on is a raspberry pi.
I am trying to build a raspberry pi cluster with 3 nodes and 1 master with kubernetes
Dose anyone have ideas on this?
Thank you all for responding to my question. I solved my problem now. For anyone who has come to my question in the future the solution was as followed.
I cloned my raspberry pi images because i wanted a basicConfig.img for when i needed to add a new node to my cluster of when one gets down.
Weave network (the plugin i used) got confused because on every node and master the os had the same machine-id. When i deleted the machine id and created a new one (and reboot the nodes) my error got fixed. The commands to do this was
sudo rm /etc/machine-id
sudo rm /var/lib/dbus/machine-id
sudo dbus-uuidgen --ensure=/etc/machine-id
Once again my patience was being tested. Because my kubernetes setup was normal and my raspberry pi os was normal. I founded this with the help of someone in the kubernetes community. This again shows us how important and great are IT community is. To the people of the future who will come to this question. I hope this solution will fix your error and will decrease the amount of time you will be searching after a stupid small thing.
Looking at the pertinent code in Kubernetes and in CNI, the specific error you see seems to indicate that it cannot find any files ending in .json, .conf or .conflist in the directory given.
This makes me think it could be something as the conf file not being present on all the hosts, so I would verify that as a first step.
I installed kubernetes on Ubuntu 16.04 (Virtualbox vm) - a single node with master tainted. It worked well. But after I restart my vm, it is not working any more.
kubectl commands are not working any more, throws this error -
The connection to the server localhost:8001 was refused - did you specify the
right host or port?
It looks similar to this thread, but the solution is not working for me.
When I try "sudo docker ps -a", all kube pods are showing in Exited status.
Any helps/pointers, please? Thanks in advance.
I've been having the same issue with my rancher 2 setup. I have two nodes in one cluster. One of my node servers was restarted and never connected to my cluster. Even though docker and containers were running fine.
One of the things i tried was reduce the number of workloads that can run in one node. I had increased it to 400. SO i put it back to 100. That's when I got my first breakthrough of what could be happening with my downed node. I go the error "Path /var/lib/docker is mounted on / but it is not a shared or slave mount." A quick search led me to a similar issue in the rancher github page. Basically a workaround by superseb fixed my issue. I sshed into my node and ran
> mount --make-rshared /
> docker start kubelet
Maybe your issue might be different, but maybe you could be having this same shared problem.