Hello,
I installed Mesosphere DC/OS in Bare Metal with 1 master, 1 agent, 1 public agent and 1 boot.
The script Ansible ran very good but i couldn't access to my the dashboard of Mesosphere.
I got an error with the mesos-dns.service :
sept. 17 13:36:17 gzlaac01 mesos-dns[14638]: 2018/09/17 13:36:17 Failed to connect to zk-5.zk:2181: dial tcp: lookup zk-5.zk on 198.51.100.3:53: no such host
sept. 17 13:36:17 gzlaac01 mesos-dns[14638]: 2018/09/17 13:36:17 Failed to connect to zk-3.zk:2181: dial tcp: lookup zk-3.zk on 198.51.100.3:53: no such host
sept. 17 13:36:18 gzlaac01 mesos-dns[14638]: 2018/09/17 13:36:18 Failed to connect to zk-4.zk:2181: dial tcp: i/o timeout
sept. 17 13:36:18 gzlaac01 mesos-dns[14638]: 2018/09/17 13:36:18 Failed to connect to zk-2.zk:2181: dial tcp: lookup zk-2.zk on 198.51.100.3:53: no such host
sept. 17 13:36:19 gzlaac01 mesos-dns[14638]: 2018/09/17 13:36:19 Failed to connect to zk-1.zk:2181: dial tcp: lookup zk-1.zk on 198.51.100.3:53: no such host
sept. 17 13:36:19 gzlaac01 mesos-dns[14638]: 2018/09/17 13:36:19 Failed to connect to zk-5.zk:2181: dial tcp: lookup zk-5.zk on 198.51.100.3:53: no such host
sept. 17 13:36:19 gzlaac01 mesos-dns[14638]: 2018/09/17 13:36:19 Failed to connect to zk-3.zk:2181: dial tcp: lookup zk-3.zk on 198.51.100.3:53: no such host
sept. 17 13:36:19 gzlaac01 mesos-dns[14638]: 2018/09/17 13:36:19 Failed to connect to zk-4.zk:2181: dial tcp: lookup zk-4.zk on 198.51.100.3:53: no such host
sept. 17 13:36:19 gzlaac01 mesos-dns[14638]: 2018/09/17 13:36:19 Failed to connect to zk-2.zk:2181: dial tcp: lookup zk-2.zk on 198.51.100.3:53: no such host
sept. 17 13:36:20 gzlaac01 mesos-dns[14638]: 2018/09/17 13:36:20 Failed to connect to zk-1.zk:2181: dial tcp: lookup zk-1.zk on 198.51.100.3:53: no such host

Can you help me please ?

inventory.py doesn't read agents private/public from GCP Terraform, only the the master and bootstrap VMs

Update docs to reflect Makefile use

Minor problem

 ___________________________________
< TASK [Test Ansible user's access] >
 -----------------------------------
        \   ^__^
         \  (oo)\_______
            (__)\       )\/\
                ||----w |
                ||     ||

fatal: [11.11.243.54 -> localhost]: FAILED! => 
{"changed": true, "cmd": "ssh centos@localhost 'sudo echo success'", 
"delta": "0:00:00.044245", "end": "2018-08-04 19:03:49.032553", 
"msg": "non-zero return code", "rc": 255, "start": "2018-08-04 19:03:48.988308", 
"stderr": "ssh: connect to host localhost port 22: Connection refused",
 "stderr_lines": ["ssh: connect to host localhost port 22: Connection refused"],
 "stdout": "", "stdout_lines": []}

When running plays/access-onprem.yml on Centos7 on Hetzner Cloud.

However note that the failure did not seem to affect the install as I did run ansible-playbook plays/install.yml just after and all is perfect

Add support for the flag master_external_loadbalancer, to avoid struggle with the DC/OS certificate.

http://docs-staging.mesosphere.com/1.11/installing/ent/custom/configuration/configuration-parameters/#master-external-loadbalancer

TF GCP dc/os agents fail, dc/sos does not get bootstrapped properly

terraform init -from-module github.com/jrx/terraform-dcos//aws/gcp overwrites main README.md with the file from github.com/jrx/terraform-dcos

Hello,
I want to ask you if i can change etc and apiserver to 1 container instead of 3 because i try to run kubernetes in a local cluster (one master, one agent) and i haven't a lot of resources.

Thank you.

Example: https://github.com/mesosphere/dcos-kubernetes-quickstart/blob/master/Makefile

Hello,

I installed Mesosphere DC/OS with this project ansible-dcos on a bare-metal cluster (boot+master+private agent+public agent) but I can't access the DC/OS dashboard from my local machine.

There is no error in the services (dcos-mesos-dns, dcos-mesos-master, dcos-admin-router...).

Any ideas ?

Document how to setup dcos cli and kubernetes for the Enterprise version

Hi there, when you try use this playbook on fresh centos 7 installation, you receive a error like that :

fatal: [192.168.0.217]: FAILED! => {"changed": false, "cmd": "yum-config-manager --add-repo https://download.docker.com/linux/centos/docker-ce.repo", "delta": "0:00:00.004238", "end": "2018-05-13 16:00:00.061159", "msg": "non-zero return code", "rc": 127, "start": "2018-05-13 16:00:00.056921", "stderr": "/bin/sh: yum-config-manager: command not found", "stderr_lines": ["/bin/sh: yum-config-manager: command not found"], "stdout": "", "stdout_lines": []}
fatal: [192.168.0.216]: FAILED! => {"changed": false, "cmd": "yum-config-manager --add-repo https://download.docker.com/linux/centos/docker-ce.repo", "delta": "0:00:00.004296", "end": "2018-05-13 16:00:00.547001", "msg": "non-zero return code", "rc": 127, "start": "2018-05-13 16:00:00.542705", "stderr": "/bin/sh: yum-config-manager: command not found", "stderr_lines": ["/bin/sh: yum-config-manager: command not found"], "stdout": "", "stdout_lines": []}

Solution :
Add in "install system utilities" task package "yum-utils" to instalation list, like bellow :

roles/common/tasks/main.yml 
- name: install system utilities
  yum:
    name: "{{ item }}"
    state: latest
  with_items:
    - tar
    - xz
    - unzip
    - curl
    - ipset
    - ntp
    - firewalld
    - dnsmasq
    - yum-utils

~/ansible-dcos$ ansible hosts.yaml -i inventory.py -m ping
[WARNING]: provided hosts list is empty, only localhost is available. Note that the implicit localhost does not match 'all'

[WARNING]: Could not match supplied host pattern, ignoring: hosts.yaml

I do not have access to Azure account, assigned to you @jrx then

When I run dcos cluster setup http://xxx.xx.xx.xx

I get the following error: URL [http://xxx.xx.xx.xx/acs/api/v1/auth/login] is unreachable.

I'm running this on an open stack cluster.

Any pointers in the right direction?

Thanks!