Closed massih10 closed 3 years ago
@massih10 have you ensured that your underlying network / firewall rules allow the necessary BGP traffic? Requirements are listed here: https://docs.projectcalico.org/getting-started/kubernetes/requirements#network-requirements
Assuming you have enabled TCP 179 in your firewall, could you also run calicoctl node status
on the second node and post it's output here?
Sorry for not replying.
The problem was with my iptables. They were messed up from too many installations of calico /Flannel / weave ...
Reset kubeadm, flush iptables (everything besides what's related to docker), re initiate kubeadm cluster and redeploy calico solved it for me.
I had this exact problem after setting up a new k8s cluster and using the Calico operator to install Calico per the instructions. None of the nodes could see any other peers. I added port 179/tcp and port 4789/udp to my firewall configuration and that resolved the issue. Might be good to include a firewall rule check in the install steps, to make sure all of the necessary ports are open, to avoid a troubleshooting exercise. I have to admit, however, that having to troubleshoot this issue caused me to learn more about how Calico works.
Hi Everyone, I'm running Kubernetes on a fresh installation of Ubuntu 20.04 Virtual Machines. The Host System is a Debian XEN providing the VMs with static IP Addresses. I initialize my Cluster on the master node with IP 192.168.220.4 via
sudo kubeadm init --apiserver-advertise-address=192.168.220.4 --pod-network-cidr=10.244.0.0/16
and then I install calico, following the docs at [https://docs.projectcalico.org/getting-started/kubernetes/self-managed-onprem/onpremises#install-calico-with-kubernetes-api-datastore-50-nodes-or-less]
curl https://docs.projectcalico.org/manifests/calico.yaml -O
kubectl apply -f calico.yaml
When its only the master node networking works fine, but as soon as I join my worker node with IP 192.168.220.8, DNS fails.
kubectl describe pod calico-node-jmwf9 -n kube-system
kubectl describe pod calico-node-8pb6d -n kube-system
kubectl get pods -A
sudo calicoctl node status
ip link
Expected Behavior
Cluster DNS should work without problem. BGP networking should automatically be established. PEER Connection shows Passive.
Current Behavior
Warning Unhealthy Number of node(s) with BGP peering established = 0 calico/node is not ready: BIRD is not ready: BGP not established with 192.168.220.8
Possible Solution
I have already manually set the right network adapter via
kubectl set env daemonset/calico-node -n kube-system IP_AUTODETECTION_METHOD=interface=eth0
Steps to Reproduce (for bugs)
sudo kubeadm init --apiserver-advertise-address=192.168.220.4 --pod-network-cidr=10.244.0.0/16
Your Environment
calicoctl version
Client Version: v3.14.0 Git commit: c97876ba Cluster Version: v3.18.1 Cluster Type: k8s,bgp,kubeadm,kddkubectl version --short
Client Version: v1.20.5 Server Version: v1.20.5lsb_release -a
No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 20.04.2 LTS Release: 20.04 Codename: focalAny help would be much appreciated.