Announcing Ansible playbooks for Ceph!

Ceph: cow based deployment with Ansible :-).

I am delighted to announce that I achieved the Ansible playbook to deploy Ceph.

I. The playbook

The playbook is available on the Ceph Github.

What does it do?

  • Authentication (cephx), this can be disabled.
  • Supports cluster public and private network.
  • Monitors deployment. You can easily start with one monitor and then progressively add new nodes. So can deploy one monitor for testing purpose. For production, I recommend to always use an odd number of monitors, 3 tends to be the standard.
  • Object Storage Daemons. Like the monitors you can start with a certain amount of nodes and then grow this number. The playbook either supports a dedicated device for storing the journal or both journal and OSD data on the same device (using a tiny partition at the beginning of the device).
  • Metadata daemons.
  • Collocation. The playbook supports collocating Monitors, OSDs and MDSs on the same machine.
  • The playbook was validated on both Debian Wheezy and Ubuntu 12.04 LTS.
  • Tested on Ceph Dumpling and Emperor.
  • A rolling upgrade playbook was written, an upgrade from Dumpling to Emperor was performed and worked.

What not supported?

II. Setup with Vagrant

I know that Vagrant has an Ansible prosionning provider, however I thought this was way to much automation. I believe that you need to put your hands on it otherwise you will not be able to perform any other deployment (specially if it is your first time with Ansible).

First modify the rc file we your home directory:

export ANSIBLE_CONFIG=<whatever_path>/.ansible.cfg

Do the same for the .ansible.cfg file:

host_key_checking = False
remote_user = vagrant
hostfile = <whatever_path>/hosts
log_path = <whatever_path>/ansible.log
ansible_managed = Ansible managed: modified on %Y-%m-%d %H:%M:%S by {uid}
private_key_file = ~/.vagrant.d/insecure_private_key
error_on_undefined_vars = False

Edit your /etc/hosts file with:

# Ansible hosts   ceph-mon0   ceph-mon1   ceph-mon2   ceph-osd0   ceph-osd1   ceph-osd2

Now since we use Vagrant and port forwarding, don’t forget to grab the SSH local port of your VMs. Then edit your hosts file accordingly.

Ok let’s get serious now. Run your virtual machines:

$ vagrant up

Test if Ansible can access the virtual machines:

$ ansible all -m ping
ceph-mon0 | success >> {
"changed": false,
"ping": "pong"

ceph-mon1 | success >> {
"changed": false,
"ping": "pong"

ceph-osd0 | success >> {
"changed": false,
"ping": "pong"

ceph-osd2 | success >> {
"changed": false,
"ping": "pong"

ceph-mon2 | success >> {
"changed": false,
"ping": "pong"

ceph-osd1 | success >> {
"changed": false,
"ping": "pong"

Ready to deploy? Let’s go!

$ ansible-playbook -f 6 -v site.yml
\ ^__^
\ (oo)\_______
(__)\ )\/\
||----w |
|| ||

ceph-mon0 : ok=13 changed=10 unreachable=0 failed=0
ceph-mon1 : ok=13 changed=9 unreachable=0 failed=0
ceph-mon2 : ok=13 changed=9 unreachable=0 failed=0
ceph-osd0 : ok=19 changed=12 unreachable=0 failed=0
ceph-osd1 : ok=19 changed=12 unreachable=0 failed=0
ceph-osd2 : ok=19 changed=12 unreachable=0 failed=0

Check the status:

$ vagrant ssh mon0 -c "sudo ceph -s"
cluster 4a158d27-f750-41d5-9e7f-26ce4c9d2d45
health HEALTH_OK
monmap e3: 3 mons at {ceph-mon0=,ceph-mon1=,ceph-mon2=}, election epoch 6, quorum 0,1,2 ceph-mon0,ceph-mon1,ceph-mon2
mdsmap e6: 1/1/1 up {0=ceph-osd0=up:active}, 2 up:standby
osdmap e10: 6 osds: 6 up, 6 in
pgmap v17: 192 pgs, 3 pools, 9470 bytes data, 21 objects
205 MB used, 29728 MB / 29933 MB avail
192 active+clean

Quick note about distro support. This playbook has been tested on both Debian Wheezy and Ubuntu.

I sincerely hope that you will enjoy this playbook :).