Is this a request for help?:
NO
Is this an ISSUE or FEATURE REQUEST? (choose one):
Issue
What version of aks-engine?:
Version: canary
GitCommit: 124379b0
GitTreeState: clean
Kubernetes version:
1.13.3
What happened:
Error registering network: failed to acquire lease: node "k8s-master-19932461-0" pod cidr not assigned
What you expected to happen:
Deploy flannel.
How to reproduce it (as minimally and precisely as possible):
Follow steps above.
Anything else we need to know:
This is a different way to enable cross OS pods communication.
馃憢 Thanks for opening your first issue here! If you're reporting a 馃悶 bug, please make sure you include steps to reproduce it.
Found the problem.
This link helps to solve it.
Flannel is usually configured on clusters that are deployed with kubeadm with a flag pod-network-cidr:
kubeadm init --pod-network-cidr=10.244.0.0/12
Now, we were able to fix the problem by patching all the nodes:
kubectl patch node <NODE_NAME> -p '{"spec":{"podCIDR":"<SUBNET>"}}'
What would be the aks-engine equivalent to set this flag?
Thanks @brusMX, it might be a good practice for us to match the kubeadm behaviour and flag nods with the podCIDR.
AKS allows for the modification of the podCidr and serviceCidr.
I was thinking of adding something like this to match what AKS does.
"masterProfile": {
"count": 3,
"dnsPrefix": "",
"vmSize": "Standard_DS2_v2",
"vnetSubnetId": "",
"vnetCidr": "",
"podCidr": "",
"serviceCidr": "",
"dnsServiceIP": "",
"dockerBridgeCidr": "172.17.0.1/32",
"availabilityProfile": "AvailabilitySet"
},
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
Most helpful comment
Thanks @brusMX, it might be a good practice for us to match the kubeadm behaviour and flag nods with the podCIDR.