Bash error during Trellis server provisioning

[okay-gray]

I new to roots and server configuration so forgive me if I seem somewhat clueless.
I’ve been following along the docs and when I go to provision my server with

ansible-playbook server.yml -e env=staging

I receive this error. (My server ip has been replaced with xx.xxx.xxx.xxx, and my domain myurl.com, where example.com appears in the error, it also appears in my actual error).

TASK [wordpress-setup : Generate self-signed certificates] *********************
skipping: [xx.xxx.xxx.xxx] => (item=myurl.com)
---------------------------------------------------
non-zero return code
/bin/bash: -c: line 0: unexpected EOF while looking for matching `)'
/bin/bash: -c: line 10: syntax error: unexpected end of file
failed: [xx.xxx.xxx.xxx] (item=no_default) => {"changed": true, "cmd": "openssl req -new -newkey rsa:2048 -days 3650 -nodes -x509 -sha256 -extensions req_ext -config <( cat <<' EOF'\n[req]\nprompt = no\ndistinguished_name = req_dn\n[req_dn]\ncommonName = example.com\n[req_ext]\nsubjectAltName = DNS:example.com\nEOF\n) -keyout no_default.key -out no_default.cert", "delta": "0:00:00.003399", "end": "2018-12-20 02:08:21.527204", "item": {"key": "no_default", "value": {"multisite": {}, "site_hosts": [{"canonical": "example.com"}], "ssl": {"enabled": true, "provider": "self-signed"}}}, "rc": 1, "start": "2018-12-20 02:08:21.523805", "stderr_lines": ["/bin/bash: -c: line 0: unexpected EOF while looking for matching `)'", "/bin/bash: -c: line 10: syntax error: unexpected end of file"], "stdout": "", "stdout_lines": []}

Any help or ideas would be appreciated.

[swalkinshaw]

never seen this before.

However, if example.com is appearing then you haven’t completely configured your wordpress_sites yet (or maybe not the proper one).

Are you editing group_vars/staging/wordpress_sites.yml? You’ll want to set your site_hosts in there according to https://roots.io/trellis/docs/wordpress-sites/#normal-settings.

[swalkinshaw]

Also, if you have a real domain/host with DNS set up, I’d suggest using the letsencrypt SSL provider rather than self-signed. That’s really only meant for development.

You can see the docs at https://roots.io/trellis/docs/ssl/#lets-encrypt

[okay-gray]

Are you editing group_vars/staging/wordpress_sites.yml?

Yes I’ve configured it accordingly and there is no instance of example.com in the file.

I’d suggest using the letsencrypt SSL provider rather than self-signed.

My /staging/wordpress_sites.yml is set to use letsencrypt, I don’t know why it’s attempting to self-sign?

In fact, here are the results of a grep in my trellis directory. (Most or all of the instances occur in comments).

$ grep -lr "example.com" ~/Sites/myurl.com/trellis -s

myurl.com/trellis/group_vars/development/vault.yml
myurl.com/trellis/group_vars/all/mail.yml
myurl.com/trellis/group_vars/all/main.yml
myurl.com/trellis/bin/deploy.sh
myurl.com/trellis/CHANGELOG.md
myurl.com/trellis/roles/letsencrypt/library/test_challenges.py
myurl.com/trellis/roles/ssmtp/defaults/main.yml
myurl.com/trellis/roles/sshd/README.md
myurl.com/trellis/roles/wordpress-setup/defaults/main.yml
myurl.com/trellis/roles/common/templates/site_hosts.j2
myurl.com/trellis/roles/nginx/templates/h5bp/directive-only/cross-domain-insecure.conf
myurl.com/trellis/README.md
myurl.com/trellis/hosts/development
myurl.com/trellis/vendor/roles/mailhog/tests/message
myurl.com/trellis/vendor/roles/mailhog/.travis.yml

and another searching for self-signed

$ grep -lr "self-signed" ~/Sites/myurl.com/trellis -s

myurl.com/trellis/group_vars/development/wordpress_sites.yml
myurl.com/trellis/roles/wordpress-setup/tasks/self-signed-certificate.yml
myurl.com/trellis/roles/wordpress-setup/tasks/main.yml
myurl.com/trellis/roles/wordpress-setup/defaults/main.yml
myurl.com/trellis/roles/wordpress-setup/templates/wordpress-site.conf.j2

[swalkinshaw]

Thank you for the great debugging and details by the way

This sounds like it could be an issue with your hosts files in Trellis. Do you happen to have the same host/IP in multiple hosts files?

If so, here’s a great explanation/solution:

Production server provision configures Staging envs if hosts are the same - Bug or Intention?

@craigpearson Thank you for reporting this! It is definitely not the intention that Trellis will load staging information when you specify env=production. It’s an ‘unexpected behavior’ resulting from an unexpected usage of including your host 195.50.84.246 in multiple environment groups (in [staging] and [production] simultaneously). Ansible sees multiple groups for the host and picks one; not always the one you want. Your unexpected usage is to have both staging and production on the same serv…

[okay-gray]

No I have the IPs for seperate DO droplets for my staging and productions hosts file.

[okay-gray]

My bad. I was running Ansible 2.8.0.dev0 . Downgrading to 2.7.5 solved my problem. Thanks for your help.

[swalkinshaw]

Did you by chance see a warning about the max tested Ansible version? It’s fine if you missed it (or ignored it ), but just wondering if its working properly.

[okay-gray]

Yeah the warning was working, just missed it.