CentOS 6 secondary container failure

For some reason the functional test has only been failing on centos-6-02 (following is prior to the commit where I changed default hostnames to include the extra hyphen)

PLAY [all] *********************************************************************

TASK [Gathering Facts] *********************************************************
Sunday 22 April 2018  05:10:55 +0000 (0:00:00.040)       0:00:00.040 ********** 
ok: [debian-stretch-test02]
ok: [debian-wheezy01]
ok: [debian-jessie02.lxc]
ok: [debian-stretch-test01]
ok: [debian-wheezy02]
ok: [fedora-2602]
ok: [fedora-2701]
fatal: [centos-602]: FAILED! => {
    "changed": false, 
    "failed": true, 
    "module_stderr": "Shared connection to centos-602 closed.\r\n", 
    "module_stdout": "", 
    "rc": 0
}

MSG:

MODULE FAILURE

ok: [centos-601]
ok: [centos-702]
ok: [debian-jessie01.lxc]
ok: [fedora-2601]
ok: [fedora-2702]
ok: [ubuntu-xenial02]
ok: [centos-701]
ok: [ubuntu-xenial01]
ok: [ubuntu-trusty02]
ok: [ubuntu-trusty01]
ok: [ubuntu-precise01]
ok: [ubuntu-precise02]
ok: [fedora-2501]
ok: [fedora-2502]

It's an itermittent issue and it turns out that the ssh_connection.retries option in Ansible's config is hardly helping. Only way to clear it out so far is the restart the build and hope for the best...maybe I should just drop it from the inventory.

lae / ansible-role-travis-lxc

CentOS 6 secondary container failure #26