Refactor running with a service to Foreman 1.22 #723
Conversation
| @@ -339,6 +339,8 @@ | |||
| timeout => 0, | |||
| } | |||
|
|
|||
| $use_foreman_service = ! $passenger | |||
There was a problem hiding this comment.
I get why we need to set passenger to false explicitly in the current setup, but I do find not setting use_foreman_service unobvious when looking at the parameters someone would be specifying. If we will reverse this very soon after the 1.22 release then I can live with it.
There was a problem hiding this comment.
Note this is currently not a parameter but rather a variable inside the body so the user can't override this. In the reverse proxy scenario I'm expanding this logic. Because I couldn't properly figure out an API, I opted not to expose this to the user directly.
| } | ||
|
|
||
| if $::foreman::use_foreman_service { | ||
| systemd::dropin_file { 'installer.conf': |
There was a problem hiding this comment.
Why is it called installer.conf ?
There was a problem hiding this comment.
I thought it'd make it easy to track where it came from. It ends up as /etc/systemd/system/foreman.service.d/installer.conf
ekohl
force-pushed
the
foreman-1.22-service
branch
from
f63b0b2 to
36501b9
Compare
ekohl
force-pushed
the
foreman-1.22-service
branch
6 times, most recently
from
02b106e to
b7b43fc
Compare
ekohl
force-pushed
the
foreman-1.22-service
branch
from
b7b43fc to
e9ffa89
Compare
|
This is now 💚 and even though the Debian side isn't merged (theforeman/foreman-packaging#3643), it will be needed on EL for 1.22 because the service is no longer present unless installed. |
|
merged, bedankt @ekohl! |
This uses the new foreman-service package. It slightly depends on the outcome of theforeman/foreman-packaging#3594.
It needs some parameters to select the port and bind address as well as tests.