kubeadm: kubeadm fails with no ipv4 address on the network interface with defaultroute

Is this a BUG REPORT or FEATURE REQUEST?

Choose one: BUG REPORT

Versions

kubeadm version (use kubeadm version): kubeadm version: &version.Info{Major:"1", Minor:"11", GitVersion:"v1.11.2", GitCommit:"bb9ffb1654d4a729bb4cec18ff088eacc153c239", GitTreeState:"clean", BuildDate:"2018-08-07T23:14:39Z", GoVersion:"go1.10.3", Compiler:"gc", Platform:"linux/amd64"}

Environment:

• Kubernetes version (use kubectl version): Client Version: version.Info{Major:"1", Minor:"11", GitVersion:"v1.11.2", GitCommit:"bb9ffb1654d4a729bb4cec18ff088eacc153c239", GitTreeState:"clean", BuildDate:"2018-08-07T23:17:28Z", GoVersion:"go1.10.3", Compiler:"gc", Platform:"linux/amd64"}

• Cloud provider or hardware configuration: Hardware

• OS (e.g. from /etc/os-release):

VERSION="7 (Core)"
ID="centos"
ID_LIKE="rhel fedora"
VERSION_ID="7"
PRETTY_NAME="CentOS Linux 7 (Core)"
ANSI_COLOR="0;31"
CPE_NAME="cpe:/o:centos:centos:7"
HOME_URL="https://www.centos.org/"
BUG_REPORT_URL="https://bugs.centos.org/"

CENTOS_MANTISBT_PROJECT="CentOS-7"
CENTOS_MANTISBT_PROJECT_VERSION="7"
REDHAT_SUPPORT_PRODUCT="centos"
REDHAT_SUPPORT_PRODUCT_VERSION="7"

• Kernel (e.g. uname -a): Linux node1-lab-a1-01 3.10.0-862.14.4.el7.x86_64 #1 SMP Wed Sep 26 15:12:11 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux

• Others: Our hosts are setup using IPv4 BGP over IPv6 (rfc5549: https://tools.ietf.org/html/rfc5549). The host ip address is attached to a loopback address and FRR bgp announces that IP to connected TOR switches (spine and leaf fabric). There is no IPv4 address on the connected interfaces, but I do have a default route that allows access to the world:

default proto bgp metric 20
	nexthop via 169.254.0.1 dev em1 weight 1 onlink
	nexthop via 169.254.0.1 dev em2 weight 1 onlink
10.101.155.0/24 proto bgp metric 20
	nexthop via 169.254.0.1 dev em1 weight 1 onlink
	nexthop via 169.254.0.1 dev em2 weight 1 onlink
10.101.246.0/24 dev em3 proto kernel scope link src 10.101.246.11
169.254.0.0/16 dev em3 scope link metric 1002
169.254.0.0/16 dev em1 scope link metric 1003
169.254.0.0/16 dev em2 scope link metric 1005
172.17.0.0/16 dev docker0 proto kernel scope link src 172.17.0.1

What happened?

Having a similar problem as #982

running

# kubeadm config images pull
unable to select an IP from default routes.

# kubeadm config images pull -v 10
I1003 20:56:35.880474   87226 interface.go:360] Looking for default routes with IPv4 addresses
I1003 20:56:35.880550   87226 interface.go:365] Default route transits interface "em1"
I1003 20:56:35.881831   87226 interface.go:174] Interface em1 is up
I1003 20:56:35.881925   87226 interface.go:222] Interface "em1" has 1 addresses :[fe80::266e:96ff:fe5f:7b48/64].
I1003 20:56:35.881957   87226 interface.go:189] Checking addr  fe80::266e:96ff:fe5f:7b48/64.
I1003 20:56:35.881989   87226 interface.go:202] fe80::266e:96ff:fe5f:7b48 is not an IPv4 address
I1003 20:56:35.882027   87226 interface.go:360] Looking for default routes with IPv6 addresses
I1003 20:56:35.882051   87226 interface.go:376] No active IP found by looking at default routes
unable to select an IP from default routes.

Same error happens with doing kubeadm init

What you expected to happen?

I expect kubeadm to work and pull the images or perform an init. Maybe have a way to specify my hosts’s ip address or interface.

How to reproduce it (as minimally and precisely as possible)?

Anything else we need to know?