TASK [kubernetes/preinstall : Hosts | populate inventory into hosts file] failed with error messasge "The task includes an option with an undefined variable. The error was: 'dict object' has no attribute 'address'"

[gashev]

Environment:

• Cloud provider or hardware configuration:
  bare metal

• OS (printf "$(uname -srm)\n$(cat /etc/os-release)\n"):

Linux 3.10.0-957.21.3.el7.x86_64 x86_64
NAME="CentOS Linux"
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"

• Version of Ansible (ansible --version):

ansible 2.7.12
  config file = /tmp/kubespray/ansible.cfg
  configured module search path = ['/tmp/kubespray/library']
  ansible python module location = /tmp/p3/lib/python3.6/site-packages/ansible
  executable location = /tmp/p3/bin/ansible
  python version = 3.6.8 (default, Aug  7 2019, 17:28:10) [GCC 4.8.5 20150623 (Red Hat 4.8.5-39)]

Kubespray version (commit) (git rev-parse --short HEAD):

8712bdd

Network plugin used:
calico

Copy of your inventory file:

[all]
node1 ansible_host=10.2.0.2 ip=10.2.0.2
node2 ansible_host=10.2.0.3 ip=10.2.0.3
node3 ansible_host=10.2.0.4 ip=10.2.0.4
node4 ansible_host=10.2.0.5 ip=10.2.0.5
node5 ansible_host=10.2.0.6 ip=10.2.0.6

[kube-master]
node1

[etcd]
node1
node2
node3

[kube-node]
node2
node3
node4
node5

[calico-rr]

[k8s-cluster:children]
kube-master
kube-node
calico-rr

Command used to invoke ansible:

ansible-playbook -i inventory/mycluster/inventory.ini --become --become-user=root cluster.yml

Output of ansible run:

TASK [kubernetes/preinstall : Hosts | populate inventory into hosts file] ************************************************************************
Saturday 28 September 2019  08:38:13 +0000 (0:00:00.697)       0:00:39.704 **** 
fatal: [node2]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'dict object' has no attribute 'address'\n\nThe error appears to have been in '/tmp/kubespray/roles/kubernetes/preinstall/tasks/0090-etchosts.yml': line 2, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n---\n- name: Hosts | populate inventory into hosts file\n  ^ here\n"}
fatal: [node3]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'dict object' has no attribute 'address'\n\nThe error appears to have been in '/tmp/kubespray/roles/kubernetes/preinstall/tasks/0090-etchosts.yml': line 2, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n---\n- name: Hosts | populate inventory into hosts file\n  ^ here\n"}
fatal: [node4]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'dict object' has no attribute 'address'\n\nThe error appears to have been in '/tmp/kubespray/roles/kubernetes/preinstall/tasks/0090-etchosts.yml': line 2, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n---\n- name: Hosts | populate inventory into hosts file\n  ^ here\n"}
fatal: [node5]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'dict object' has no attribute 'address'\n\nThe error appears to have been in '/tmp/kubespray/roles/kubernetes/preinstall/tasks/0090-etchosts.yml': line 2, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n---\n- name: Hosts | populate inventory into hosts file\n  ^ here\n"}
fatal: [node1]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'dict object' has no attribute 'address'\n\nThe error appears to have been in '/tmp/kubespray/roles/kubernetes/preinstall/tasks/0090-etchosts.yml': line 2, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n---\n- name: Hosts | populate inventory into hosts file\n  ^ here\n"}

Anything else do we need to know:

[paunmihai]

Any update on this? I face the same issue when I'm trying to install on CentOS 7.5.1804 running in a VirtualBox VM.

[Docteur-RS]

Duplicate of 4750 I guess...
A true fix would be great. If its even doable...

[champtar]

https://github.com/kubernetes-sigs/kubespray/blob/master/roles/kubespray-defaults/defaults/main.yaml#L393

# Set 127.0.0.1 as fallback IP if we do not have host facts for host
fallback_ips_base: |
  ---
  {% for item in groups['k8s-cluster'] + groups['etcd'] + groups['calico-rr']|default([])|unique %}
  {{ item }}: "{{ hostvars[item].get('ansible_default_ipv4', {'address': '127.0.0.1'})['address'] }}"
  {% endfor %}
fallback_ips: "{{ fallback_ips_base | from_yaml }}"

So in some cases, ansible_default_ipv4 is present but doesn't contains address, as show by adding a debug

TASK [kubernetes/preinstall : debug] *********************************************************************************************************************************************************************************************************
Wednesday 22 January 2020  16:50:24 -0500 (0:00:01.189)       0:01:15.581 ***** 
ok: [etienne-ks141] => {
    "ansible_default_ipv4": {}
}

After some testing, it seems ansible look at the source address that the host would use to contact 8.8.8.8
To test it:

ip link add dummy0 type dummy
ip addr add 8.8.8.8/32 dev dummy0
or
ip r add blackhole 8.8.8.8
or
by not having a default route ;)

$ ansible -i <ip>, -m setup <ip> | grep ansible_default_ipv4
        "ansible_default_ipv4": {},

As suggested here
https://medium.com/opsops/ansible-default-ipv4-is-not-what-you-think-edb8ab154b10
a workaround for ansible_default_ipv4 could be

ansible_default_ipv4.address|default(ansible_all_ipv4_addresses[0])

If ansible_default_ipv4 is always present, another fix is (not tested)

hostvars[item]['ansible_default_ipv4'].get('address', '127.0.0.1')

[champtar]

And just read #5394 that explain just what I wrote ...

[cferrera]

Any update on this?

I have the same issue.

• Azure
• CentOS 7
• ansible 2.9.6
• KubeSpray 2.12.3
• Weave
• 3 master; 3 node;

[champtar]

#5394 is now merged, @gashev is this fixed for you ?

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

[bmelbourne]

@gashev @cferrera
PR #5394 may have fixed this issue and will be included in v2.12.4+ releases.

Can you confirm?

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

[k8s-ci-robot]

@fejta-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.