Update to newest Digital Ocean modules

[w3irdrobot]

It appears that Ansible has created a new set of modules that no longer rely on dopy and add more functionality. The current one used in the book will be deprecated in 1.12. It may be beneficial to move to using the newer module(s).

[bodzio4749]

Yeap, I wanted to use my $100 for two months and failed miserably, so I tried to run this project. I failed in that too. The message was about the deprecated module. I guess, Jeff is going to be very busy fixing the code, rewriting the book, and making plans for another Ansible 101 episode. :) In the meantime, I have 8 days to use $87 in credit on DigitalOcean.

[geerlingguy]

@SearsAW - Yeah that chapter's on my list soon, after I finish some new work on the Tower/AWX chapter. I've been working my way through the book trying to make notes on examples that need revision/rewrites as I've gone through Ansible 101.

[geerlingguy]

There are two examples I need to update:

• Chapter 8 - DigitalOcean Dynamic Inventory: https://github.com/geerlingguy/ansible-for-devops/tree/master/dynamic-inventory/digitalocean
• Chapter 9 - LAMP HA Infrastructure, DigitalOcean Provisioner: https://github.com/geerlingguy/ansible-for-devops/blob/master/lamp-infrastructure/provisioners/digitalocean.yml

[geerlingguy]

Finished updating the first example. Working on the second.

[geerlingguy]

The chapter 9 example is all working... but I'm getting this from the webservers:

Host 104.131.49.132
MySQL Connection (104.131.51.172): FAIL
MySQL Connection (104.131.51.195): FAIL
Memcached Connection: FAIL

[geerlingguy]

I can connect to both MySQL servers from the CLI mysql without an issue from the Apache servers... so maybe a PHP issue?

[geerlingguy]

So... that was strange: both using PHP on the CLI php index.php and using telnet or mysql on the CLI allowed me to connect to MySQL and Memcached, when run under the root user.

But when I tried loading the web page (via Apache / mod_php), it would not succeed. No errors that I could see or messages in secure log. I disabled SELinux (setenforce 0) and then everything worked... so apparently it was blocking HTTPD and so I had to set:

# setsebool -P httpd_can_network_memcache 1
# setsebool -P httpd_can_network_connect_db 1

Or in Ansible:

    - name: Configure SELinux to allow HTTPD connections.
      seboolean:
        name: "{{ item }}"
        state: true
        persistent: true
      with_items:
        - httpd_can_network_connect_db
        - httpd_can_network_memcache

Testing this now.

[geerlingguy]

I got things working, but now replication is not starting on the servers... maybe a bug in my MySQL role?

Host 104.131.38.92
MySQL Connection (104.131.38.165): PASS (slave)
MySQL Connection (104.131.38.206): PASS (slave)
Memcached Connection: PASS

[geerlingguy]

Yay after the above fix:

Host 104.131.21.40
MySQL Connection (104.131.21.143): PASS (master)
MySQL Connection (104.131.13.87): PASS (slave)
Memcached Connection: PASS

I also had to use the chomping folded scalar to get rid of a newline character:

mysql_replication_role: >-
  {{ 'master' if (item.1.data.droplet.name == 'a4d.lamp.db.1')
  else 'slave' }}