Go to file
Pierre-Yves Chibon a11aa065ad Start working on the darkserver role 2014-12-06 20:33:47 +01:00
callback_plugins Catch and ignore fedmsg warnings about multiple initializations. 2014-06-30 14:44:52 +00:00
files [copr] backend config: added spawn_in_advance option, backported for host based version 2014-11-26 15:09:48 +01:00
handlers Restart fcomm-cache-worker when config changes. 2014-11-17 15:35:29 +00:00
inventory Commit fas server ansible playbook from ticket 4394 2014-12-06 18:33:44 +00:00
library Drop local yum ansible module to test something. 2014-03-12 17:20:41 +00:00
playbooks Fix things that are now roles 2014-12-06 18:45:13 +00:00
roles Start working on the darkserver role 2014-12-06 20:33:47 +01:00
scripts Publish fedoratagger db dump. 2014-10-27 17:17:06 +00:00
tasks Remove all accelerate-mode knobs and toggles now that pipelining is a go. 2014-11-12 19:15:06 +00:00
vars Remove all accelerate-mode knobs and toggles now that pipelining is a go. 2014-11-12 19:15:06 +00:00
.gitignore Migrate a bunch of things to roles. Thanks to misc! 2013-08-19 20:12:26 +00:00
CONVENTIONS Clean up some more old syntax 2014-01-08 17:42:18 +00:00
README Update keypair used in example host vars 2014-08-09 17:53:42 +00:00
TODO restructure virt to be more like cloud creation 2013-05-03 16:56:38 +00:00
master.yml Add fas to master.yml 2014-12-06 18:34:38 +00:00

README

== ansible repository/structure ==

files - files and templates for use in playbooks/tasks
      - subdirs for specific tasks/dirs highly recommended

inventory - where the inventory and additional vars is stored
          - All files in this directory in ini format 
          - added together for total inventory
  group_vars: 
          - per group variables set here in a file per group 
  host_vars: 
          - per host variables set here in a file per host 

library - library of custom local ansible modules

playbooks - collections of plays we want to run on systems

tasks - snippets of tasks that should be included in plays

roles - specific roles to be use in playbooks. 
        Each role has it's own files/templates/vars

== Paths ==

public path for everything is:

 /srv/web/infra/ansible

private path - which is sysadmin-main accessible only is:

 /srv/private/ansible

In general to run any ansible playbook you will want to run:

sudo -i ansible-playbook /path/to/playbook.yml

== Cloud information ==

cloud instances:
to startup a new cloud instance and configure for basic server use run (as
root):

el6:
sudo -i ansible-playbook /srv/web/infra/ansible/playbooks/el6_temp_instance.yml

f19:
sudo -i ansible-playbook /srv/web/infra/ansible/playbooks/f19_temp_instance.yml


The -i is important - ansible's tools need access to root's sshagent as well
as the cloud credentials to run the above playbooks successfully.

This will setup a new instance, provision it and email sysadmin-main that
the instance was created, it's instance id (for terminating it, attaching
volumes, etc) and it's ip address.

You will then be able to login, as root.

You can add various extra vars to the above commands to change the instance
you've just spun up.

variables to define:
instance_type=c1.medium
security_group=default
root_auth_users='username1 username2 @groupname'
hostbase=basename for hostname - will have instance id appended to it


define these with:

--extra-vars="varname=value varname1=value varname2=value"

Name        Memory_MB  Disk   VCPUs
m1.tiny     512        0      1    
m1.small    2048       20     1    
m1.medium   4096       40     2    
m1.large    8192       80     4    
m1.xlarge   16384      160    8    
m1.builder  5120       50     3    

Setting up a new persistent cloud host:
1. select an ip:
   source /srv/private/ansible/files/openstack/persistent-admin/ec2rc.sh
   oeuca-describe-addresses
  - pick an ip from the list that is not assigned anywhere
  - add it into dns - normally in the cloud.fedoraproject.org but it doesn't
    have to be

2. If needed create a persistent storage disk for the instance:
   source /srv/private/ansible/files/openstack/persistent-admin/ec2rc.sh
   euca-create-volume -z nova -s <size in gigabytes>


3. set up the host/ip in ansible host inventory
   - add to ansible/inventory/inventory under [persistent-cloud]
     - either the ip itself or the hostname you want to refer to it as

4. setup the host_vars
   - create file named by the hostname or ip you used in the inventory
   - for adding persistent volumes add an entry like this into the host_vars file

   volumes: ['-d /dev/vdb vol-BCA33FCD', '-d /dev/vdc vol-DC833F48']

   for each volume you want to attach to the instance.

   The device names matter - they start at /dev/vdb and increment. However,
   they are not reliable IN the instance. You should find the device, partition
   it, format it and label the formatted device then mount the device by label
   or  by UUID. Do not count on the device name being the same each time.


Contents should look like this (remove all the comments)

---
# 2cpus, 3GB of ram 20GB of ephemeral space
instance_type: m1.large 
# image id - see global vars. You can also use euca-describe-images to find other images as well
image: "{{ el6_qcow_id }}"
keypair: fedora-admin-20130801
# what security group to add the host to
security_group: webserver 
zone: fedoracloud 
# instance id will be appended
hostbase: hostname_base- 
# ip should be in the 209.132.184.XXX range
public_ip: $ip_you_selected
# users/groups who should have root ssh access
root_auth_users:  skvidal bkabrda 
description: some description so someone else can know what this is

The available images can be found by running::
   source /srv/private/ansible/files/openstack/persistent-admin/ec2rc.sh
   euca-describe-images | grep ami

4. setup a host playbook ansible/playbooks/hosts/$YOUR_HOSTNAME_HERE.yml
   Note: the name of this file doesn't really matter but it should normally
         be the hostname of the host you're setting up. 

- name: check/create instance
  hosts: $YOUR_HOSTNAME/IP HERE
  user: root
  gather_facts: False

  vars_files:
   - /srv/web/infra/ansible/vars/global.yml
   - "{{ private }}/vars.yml"

  tasks:
  - include: "{{ tasks }}/persistent_cloud.yml"

- name: provision instance
  hosts: $YOUR_HOSTNAME/IP HERE
  user: root
  gather_facts: True

  vars_files:
   - /srv/web/infra/ansible/vars/global.yml
   - "{{ private }}/vars.yml"
   - /srv/web/infra/ansible/vars//{{ ansible_distribution }}.yml

  tasks:
  - include: "{{ tasks }}/cloud_setup_basic.yml
  # fill in other actions/includes/etc here

  handlers:
  - include: "{{ handlers }}/restart_services.yml


5. add/commit the above to the git repo and push your changes


6. set it up:
   sudo -i ansible-playbook /srv/web/infra/ansible/playbooks/hosts/$YOUR_HOSTNAME_HERE.yml

7. login, etc

You should be able to run that playbook over and over again safely, it will
only setup/create a new instance if the ip is not up/responding.

SECURITY GROUPS
- to edit security groups you must either have your own cloud account or
  be a member of sysadmin-main

This gives you the credential to change things in the persistent tenant
- source /srv/private/ansible/files/openstack/persistent-admin/ec2rc.sh


This lists all security groups in that tenant:
- euca-describe-groups | grep GROUP

the output will look like this:
euca-describe-groups  | grep GROU
GROUP	d4e664a10e2c4210839150be09c46e5e	default	default
GROUP	d4e664a10e2c4210839150be09c46e5e	jenkins	jenkins instance group
GROUP	d4e664a10e2c4210839150be09c46e5e	logstash	logstash security group
GROUP	d4e664a10e2c4210839150be09c46e5e	smtpserver	list server group. needs web and smtp
GROUP	d4e664a10e2c4210839150be09c46e5e	webserver	webserver security group
GROUP	d4e664a10e2c4210839150be09c46e5e	wideopen	wideopen


This lets you list the rules in a specific group:
- euca-describe-group groupname

the output will look like this:

euca-describe-group wideopen
GROUP	d4e664a10e2c4210839150be09c46e5e	wideopen	wideopen
PERMISSION	d4e664a10e2c4210839150be09c46e5e	wideopen ALLOWS	tcp	1	65535	FROM	CIDR	0.0.0.0/0
PERMISSION	d4e664a10e2c4210839150be09c46e5e	wideopen ALLOWS	icmp	-1	-1	FROM	CIDR	0.0.0.0/0


To create a new group:
euca-create-group -d "group description here" groupname

To add a rule to a group:
euca-authorize -P tcp -p 22 groupname
euca-authorize -P icmp -t -1:-1 groupname

To delete a rule from a group:
euca-revoke -P tcp -p 22 groupname

Notes:
- Be careful removing or adding rules to existing groups b/c you could be
impacting other instances using that security group.

- You will almost always want to allow 22/tcp (sshd) and icmp -1 -1 (ping
and traceroute and friends). 




TERMINATING INSTANCES

For transient:
1. source /srv/private/ansible/files/openstack/transient-admin/ec2rc.sh

  - OR -

For persistent:
1. source /srv/private/ansible/files/openstack/persistent-admin/ec2rc.sh

2. euca-describe-instances | grep <ip of your instance>

3. euca-terminate-instances <the id, something like i-00000295>