Hello Tyler,
> I have gotten Foreman 1.5 working on CentOS 6.5 standalone with it's own
> Puppet installation however, once I start trying to stand it up with
> pe-puppet already installed, I encounter numerous issues that I have
> seemingly fixed but have finally ran into one where I have no clue where to
> go to fix.
Is this new installation or upgrade?
> OS: CentOS 6.5
> Foreman Version: 1.5
> Puppet Version: 3.4.3 (Puppet Enterprise 3.2.2)
> Ruby Version: 1.8.7
Hmm from what I know, Puppet 3.4 is not supported on Ruby 1.8.7. When
you run our installer, does it use Ruby 1.8 or do you have some kind of
side-installation of Ruby 1.9+ for Puppet 3.4?
> Puppet::Type::Foreman_smartproxy::ProviderRest: feature foreman_api is
> missing
We have switched over from foreman_api to apipie-bindings gem which uses
V2 API and allows us also to define HTTP request timeouts. Can you check
if you have apipie installed?
rpm -qa|grep apipie
It should pick it up, and only if it was not found, it should carry on
with the old provider.
> Something that might be related (or possibly just what the installer itself
> is seeing) is that if outside the installer I do service foreman-proxy
> start it shows successful with exit code of 0, however stopping the
> service:
>
> [root@hostname.p ~] # service foreman-proxy start
> Starting foreman-proxy: [ OK ]
> [root@hostname.p ~] # echo $?
> 0
> [root@hostname.p ~] # service foreman-proxy stop
> Stopping foreman-proxy: foreman-proxy was not running. [FAILED]
> [root@hostname.p ~] # echo $?
> 0
The question now is - was foreman-proxy really starting up? Do you see
it in the list of processes?
> I have looked up the missing foreman_api issue and came across a similar
> thread here that seems to have been abandoned. I tried a suggestion found
> there of installing the foreman_api package before installing Puppet to no
> avail.
Hmmm please check which version of Ruby and Puppet our installer use,
because looks like you maybe installed another Ruby version with
different gemset where there is no foreman_api or apipie gem.
> [root@hostname.p ~] # rpm -qa | grep foreman_api
> rubygem-foreman_api-0.1.11-1.el6.noarch
This is maybe not being picked up by the other Ruby installation ^^^
> Before arriving at this point I encountered ERRORs during installation that
> involved postgresql locking up and unable to start, this was resolved by
> killing the postmaster process and that part of the installation seems to
> work fine now.
I have never heard of such an error.
>
> I encountered another similar bug with httpd which also was fixed by
> killing stuck httpd processes.
Again, I have never heard of such an error. Is your OS installation
correct? No hardware issues? 
> I also encountered that nasty Ruby segfault error going around which is
> fixed by placing GC.disable at the top of the /usr/bin/puppet file
Yeah, this one is PITA. This reveals you are/were using 1.8 version.
···
> Any help or pointers as to what I might be doing wrong or what I can do to > get these two playing nice would be much appreciated. I know this is a > huge info dump and appreciate anyone willing to help as I am now officially > stuck.–
Later,
Lukas “lzap” Zapletal
irc: lzap #theforeman