Share multiple node on openshift origin inventory

Question

I'm trying to configure my inventory so I can install a 3 Master, 3 etcd, 1 compute and 1 infra node. The thing is that I trying to do it all in 3 servers. So the need to share them.

I also have a 4 server that I'm using as LB.

This is the host file:

[masters]
server1
server2
server3

[etcd]
server1
server2
server3

[nodes]
server1 openshift_node_group_name="node-config-master"
server2 openshift_node_group_name="node-config-all-in-one"
server3 openshift_node_group_name="node-config-master-infra"

[lb]
server4

[OSEv3:children]
masters
nodes
etcd
lb

[OSEv3:vars]
ansible_user=dxcooper
ansible_become=yes

openshift_deployment_type=origin
openshift_release="3.11"
openshift_master_default_subdomain=apps.<subdomains>
openshift_master_cluster_hostname=server4
openshift_disable_check=docker_storage

debug_level=2

openshift_http_proxy=http://<proxy>
openshift_https_proxy=http://<proxy>
openshift_no_proxy='localhost,127.0.0.0,127.0.0.1,127.0.1.1,local.home,.svc,.local,10.87.32.0/19,.<subdomains>'

The prerequisites.yml runs without problem.
When I run the deploy_cluster.yml, during the etcd installation it fails during the certificate creation part.

Any idea what I'm missing?

Hi there! What error do you got? Also, and only for not assuming the answer, your VM CIDR is 10.87.32.0/19? — Armando Cuevas
It was a proxy problem. The openshift_no_proxy needs to have the IP as well as the FQDN. — radicaled

radicaled radicaled · Accepted Answer · 2020-04-14T21:07:03

I have been looking around and I saw several people with the same problem as well as bug in redhat.

For me it was a proxy problem.
The openshift_no_proxy needs to have the IP as well as the FQDN.

Share multiple node on openshift origin inventory

1 Answers