My understanding is that the Puppet CLI parameters will be the values taken in at run time.
I have two code directories; test and production.
They are exact replicas of each other.
If I force the puppetserver to run /etc/puppetlabs/test -- it will run without issue and as expected.
When I run the command;
puppet agent -t --codedir /etc/puppetlabs/test --noop
It will still run code out of /etc/puppetlabs/production. I tested this by putting a notify statement in production, and not having one in test.
Every time I ran it, it ran with the notify statement indicating it was still reading /etc/puppetlabs/production.
Does anyone have any idea why it would not be using the parameter I input?
The puppetmaster does have the master.code.dir commented out.
The puppet.conf file on master only has an agent environment
I'm running this from a node (different server). I verified the puppet.conf for the server I'm running it from also does not have values, though at one point of testing I explicitly hardcoded them and it still failed.
Ideas, thoughts?