Upstream members are never collected

[linuxdaemon]

Affected Puppet, Ruby, OS and module versions/distributions

• Puppet: 4.10.10
• Ruby: 2.1.9p490 (2016-03-30 revision 54437) [x86_64-linux]
• Distribution: Ubuntu 16.04
• Module version: 0.15.0

How to reproduce (e.g Puppet code you use)

node1.example.com

::nginx::resource::upstream { 'test_upstream': }

# Some server or location which proxies to `http://test_upstream`

node2.example.com

@@::nginx::resource::upstream::member {$fqdn:
  upstream => 'test_upstream',
  server => $::ipaddress,
  port => 80
}

What are you seeing

Deployment fails on node2 due to the upstream having no members specified.

What behaviour did you expect instead

The upstream should collect all exported members when members is not set explicitly.

Output log

Info: Using configured environment 'production'
Info: Retrieving pluginfacts
Info: Retrieving plugin
Info: Loading facts
Info: Caching catalog for bnc.snoonet.org
Info: Applying configuration version '1541692371'
Notice: /Stage[main]/Snoonet::Web::Sites::Anope/Nginx::Resource::Upstream[anope_servers]/Concat[/etc/nginx/conf.d/anope_servers-upstream.conf]/File[/etc/nginx/conf.d/anope_servers-upstream.conf]/ensure: defined content as '{md5}a20d79a0d0491d83667ee63e43648fad'
Info: Concat[/etc/nginx/conf.d/anope_servers-upstream.conf]: Scheduling refresh of Class[Nginx::Service]
Notice: /Stage[main]/Snoonet::Web::Sites::Anope/Nginx::Resource::Server[anope]/File[anope.conf symlink]/ensure: created
Info: /Stage[main]/Snoonet::Web::Sites::Anope/Nginx::Resource::Server[anope]/File[anope.conf symlink]: Scheduling refresh of Class[Nginx::Service]
Info: Class[Nginx::Service]: Scheduling refresh of Service[nginx]
Error: /Stage[main]/Nginx::Service/Service[nginx]: Failed to call refresh: Systemd restart for nginx failed!
journalctl log for nginx:
-- Logs begin at Sat 2018-11-03 06:40:01 MDT, end at Thu 2018-11-08 08:52:56 MST. --
Nov 08 08:52:55 bnc.snoonet.org systemd[1]: Stopping A high performance web server and a reverse proxy server...
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: Stopped A high performance web server and a reverse proxy server.
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: Starting A high performance web server and a reverse proxy server...
Nov 08 08:52:56 bnc.snoonet.org nginx[17734]: nginx: [emerg] no servers are inside upstream in /etc/nginx/conf.d/anope_servers-upstream.conf:3
Nov 08 08:52:56 bnc.snoonet.org nginx[17734]: nginx: configuration file /etc/nginx/nginx.conf test failed
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: nginx.service: Control process exited, code=exited status=1
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: Failed to start A high performance web server and a reverse proxy server.
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: nginx.service: Unit entered failed state.
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: nginx.service: Failed with result 'exit-code'.

Error: /Stage[main]/Nginx::Service/Service[nginx]: Systemd restart for nginx failed!
journalctl log for nginx:
-- Logs begin at Sat 2018-11-03 06:40:01 MDT, end at Thu 2018-11-08 08:52:56 MST. --
Nov 08 08:52:55 bnc.snoonet.org systemd[1]: Stopping A high performance web server and a reverse proxy server...
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: Stopped A high performance web server and a reverse proxy server.
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: Starting A high performance web server and a reverse proxy server...
Nov 08 08:52:56 bnc.snoonet.org nginx[17734]: nginx: [emerg] no servers are inside upstream in /etc/nginx/conf.d/anope_servers-upstream.conf:3
Nov 08 08:52:56 bnc.snoonet.org nginx[17734]: nginx: configuration file /etc/nginx/nginx.conf test failed
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: nginx.service: Control process exited, code=exited status=1
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: Failed to start A high performance web server and a reverse proxy server.
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: nginx.service: Unit entered failed state.
Nov 08 08:52:56 bnc.snoonet.org systemd[1]: nginx.service: Failed with result 'exit-code'.

Notice: Applied catalog in 2.60 seconds

Any additional information you'd like to impart

The issue appears to be that the members parameter on the upstream resource is set to a default of an empty Hash but

puppet-nginx/manifests/resource/upstream.pp

Line 137 in ae2af80

if $members != undef {

checks against undef, which the value can never be as setting members to undef in a resource declaration results in it using the default value.