Flannel Bare Metal pods fails!

[lmyint55]

My Ubuntu 16.04 bare metal servers can't communicate to each other. I have different addresses/subnets for the Ubuntu server and Kubernetes/flannel.

Expected Behavior

Flannel Pods are no running on all of the servers!

Current Behavior

THIS IS MY MASTER SERVER:

lmyint45@vadmin-kubernetes-001:/etc/network$ kubectl get pods -n kube-system
NAME                                            READY     STATUS             RESTARTS   AGE
etcd-vadmin-kubernetes-001                      1/1       Running            1          21h
kube-apiserver-vadmin-kubernetes-001            1/1       Running            1          21h
kube-controller-manager-vadmin-kubernetes-001   1/1       Running            1          21h
kube-dns-86f4d74b45-v8bhl                       3/3       Running            3          21h
kube-flannel-ds-9zvfg                           0/1       CrashLoopBackOff   237        19h
kube-flannel-ds-rqbkh                           0/1       CrashLoopBackOff   215        17h
kube-flannel-ds-rqp29                           0/1       CrashLoopBackOff   60         4h
kube-flannel-ds-w564z                           1/1       Running            1          21h
kube-flannel-ds-xnrmk                           0/1       CrashLoopBackOff   250        20h
kube-proxy-5gjtd                                1/1       Running            1          19h
kube-proxy-64mjj                                1/1       Running            1          21h
kube-proxy-84rsh                                1/1       Running            0          4h
kube-proxy-hswl4                                1/1       Running            1          17h
kube-proxy-wgsvr                                1/1       Running            2          20h
kube-scheduler-vadmin-kubernetes-001            1/1       Running            1          21h
====================================================================

lmyint45@vadmin-kubernetes-001:/etc/network$ kubectl logs kube-flannel-ds-9zvfg -n kube-system
I0330 19:17:00.440707       1 main.go:474] Determining IP address of default interface
I0330 19:17:00.441208       1 main.go:487] Using interface with name enp4s0f0 and address 10.160.16.103
I0330 19:17:00.441232       1 main.go:504] Defaulting external address to interface address (10.160.16.103)
I0330 19:17:00.484948       1 kube.go:130] Waiting 10m0s for node controller to sync
I0330 19:17:00.487781       1 kube.go:283] Starting kube subnet manager
I0330 19:17:01.488074       1 kube.go:137] Node controller sync successful
I0330 19:17:01.488161       1 main.go:234] Created subnet manager: Kubernetes Subnet Manager - vadmin-kubernetes-003
I0330 19:17:01.488183       1 main.go:237] Installing signal handlers
I0330 19:17:01.488489       1 main.go:352] Found network config - Backend type: vxlan
I0330 19:17:01.488693       1 vxlan.go:119] VXLAN config: VNI=1 Port=0 GBP=false DirectRouting=false
E0330 19:17:01.489228       1 main.go:279] Error registering network: failed to acquire lease: node "vadmin-kubernetes-003" pod cidr not assigned
I0330 19:17:01.489329       1 main.go:332] Stopping shutdownHandler...
=====================================================================

Ubuntu Network is on 10.160.16.0/23 and flannel is setup on 10.160.100.0/24.
They are on blade servers with a switch backplane.
how do I make them communicate with each other.
There's no calico etc., just plain vanilla flannel.
please help.

So there's a problem with

Possible Solution

Steps to Reproduce (for bugs)

Context

We are testing to deploy kubernetes on-prem and will migrate current Docker containers to it.

Your Environment

• Calico version: none
• Flannel version: v0.9.1-amd64
• Orchestrator version: v1.10.0
• Operating System and version: Ubuntu 16.04 LTS
• Link to your project (optional):

[caseydavenport]

@lmyint55 if this is just a pure flannel question, could you please re-open the issue against the flannel repo? https://github.com/coreos/flannel/issues