nzwulfin

Ansible scaffolding for project, created in root of project directory

$ ansible-galaxy init ansible
$ mkdir -p ansible/playbooks/groups ansible/playbooks/hosts ansible/roles

Create the base role for the project for all common tasks 

$ ansible-galaxy init -p ansible/roles/base

Project should now have TLD of ansible that holds the playbooks and roles

nzwulfin

Tunir VM
Create F23 virtual machine (2G, 2vC, 10G)
u: tunir / p: Ap
git clone https://github.com/kushaldas/tunir.git
install python-paramiko python-sqlalchemy python-redis redis python-libcloud
redis (no config, start / enable)
git clone https://github.com/nzwulfin/tunirtests.git
fedora to remote
cp fedora.txt .json ~/tunir
 * curl -O http://192.168.122.246:8000/tunirtests.tar.gz

nzwulfin

## Set up new branch based on existing tree and deploy
sudo ostree commit -b synthetic_test --tree=ref=fedora-atomic:fedora-atomic/24/x86_64/docker-host | tee target_commit.txt
sudo ostree admin deploy synthetic_test
sudo systemctl reboot
atomic host status
<active commit == target_commit.txt>

## Create a 'noop' commit based on the new branch as an upgrade target
sudo ostree commit -b synthetic_test --tree=ref=synthetic_test | tee target_upgrade.txt
sudo atomic host upgrade

nzwulfin

I chatted with miabbott on IRC to find the right `ostree rev-parse` syntax to find the parent of the current deployment:
  `ostree rev-parse fedora-atomic/24/x86_64/docker-host^`
This would allow the test case to find the previous tree to deploy to test rollbacks and upgrades.

Building test cases for Tunir, I wanted to see what the error output would be for the very first tree in a release cycle (i.e. Fedora 24) so the test case wouldn't try to rollback to something that didn't exist.

After updating to the current tree (24.45) I then deployed 24, and rebooted.  Once on 24, I rolled back to 24.45.  The oddness happened here.  I tried the `ostree rev-parse` command on this rolled back 24.45 and it threw the error as if the tree was still 24.

If I ran an upgrade from 24 to 24.45, the `ostree rev-parse` worked as expected and showed the parent, 24.44.

nzwulfin

sudo atomic host status

sudo atomic host deploy $(ostree rev-parse fedora-atomic/24/x86_64/docker-host^)

sudo systemctl reboot

sudo atomic host status

nzwulfin

docker run -it --name target centos:7

ansible-playbook -i vms site.yml 

nzwulfin

import requests, os

USER='nzwulfin'
#API_TOKEN=
GIT_API_URL='https://api.github.com'

api = GIT_API_URL + '/users/' + USER + '/repos?type=all'

def get_repos(url):
    try:

nzwulfin

rpm-ostree compose tree --proxy=http://127.0.0.1:8123 --repo=/srv/rpm-ostree/fedora-atomic/23 fedora-atomic-docker-host.json

sudo ostree remote add atomic-storage-clients http://192.168.122.62 --no-gpg-verify

sudo atomic host rebase atomic-storage-clients:fedora-atomic/rawhide/x86_64/docker-host
  fedora-atomic/rawhide/x86_64/atomic-storage-clients-host

CEPH (on ceph-admin):
mkdir test-cluster && cd test-cluster
ceph-deploy new ceph-server-1 ceph-server-2 ceph-server-3

nzwulfin

net-dchp-leases:

 Expiry Time          MAC address        Protocol  IP address                Hostname        Client ID or DUID
-------------------------------------------------------------------------------------------------------------------
 2016-01-22 11:20:50  52:54:00:8f:f2:d5  ipv4      192.168.122.62/24         composer        ff:00:8f:f2:d5:00:01:00:01:1e:04:85:9c:52:54:00:8f:f2:d5
 2016-01-22 11:36:10  52:54:00:b9:5e:96  ipv4      192.168.122.126/24        atomic-001      ff:00:b9:5e:96:00:01:00:01:1e:34:fb:ee:52:54:00:b9:5e:96

sudo virsh domifaddr fedora23-composer
 Name       MAC address          Protocol     Address
-------------------------------------------------------------------------------

nzwulfin

Test steps for atomic

Below are some steps to roughly test an atomic host from Project Atomic.

Booting with cloud-init

First step is to start an atomic host using any method/cloud provider