Openshift-ansible: FAILED - RETRYING: HANDLER: openshift_master: Verify API Server (120 tries left)

[Deepak-Vohra]

Getting message (repeated 120 times)
FAILED - RETRYING: HANDLER: openshift_master: Verify API Server (120 tries left)

And subsequently:
FAILED! => {"changed": false, "cmd": ["curl", "--silent", "--cacert", "/etc/origin/master/ca-bundle.crt", "https://ec2-54-81-143-129.compute-1.amazonaws.com:8443/healthz/ready"], "delta": "0:00:05.600946", "end": "2016-08-08 22:34:36.218520", "failed": true, "rc": 51, "start": "2016-08-08 22:34:30.617574", "stderr": "", "stdout": "", "stdout_lines": [], "warnings": ["Consider using get_url or uri module rather than running curl"]}

[Deepak-Vohra]

Seems to be because of DNS not being configured.

[Deepak-Vohra]

The NetworkManager was to be installed. Issue is fixed and may be closed, but refer the request to install NetworkManager automatically.
https://github.com/openshift/openshift-ansible/issues/2272

[lannyMa]

sorry i cannnot unterstand

[poltorres7]

Hello I have the same error "Verify API server"
I have installed the NetworkManager but it still failing.

Does any body have found a solution?

[samvarankashyap]

Facing the same issue ...
Anyone anyother solution

[rushabh268]

@abutcher Seeing the same issue on Atomic cluster where NetworkManager is installed on the master node

[poltorres7]

Hi @samvarankashyap & @rushabh268 I've solve it. The problem was that I had a proxy configurated and openshift was going out in that route, when I disable it, the request was resolved it.

[ben-le]

@poltorres7; that's meant your cluster didn't include the HAProxy; How to resolve this issue with the HAProxy configuration in the HA Cluster?

[tusharsanas]

Hi,

I am facing the same issues FAILED - RETRYING: HANDLER: openshift_master: Verify API Server (120 tries left)

Few details of my setup:
3 Master, 3 Compute, 3 Infra and 3 GlusterFS.
1 External LB and another 1 Internal LB

I did check with below curl command
curl -k -vvv https://master002-internal.amosdemo.io:8443/healthz/ready

But my master is not able to connect internal load balancer.
My internal load balancer is network LB, and all the master instances attached to internal LB are healthy.

Please can anyone help me to resolve above issue.

[ben-le]

@tusharsanas
Make sure the LB hostname is correct in the inventory file, the master /etc/hosts, and the DNS records and then uninstall the OpenShift cluster
ansible-playbook -i openshift-hosts playbooks/adhoc/uninstall.yml
After uninstall completed and run install playbook again
Hope this helps