(PE-37235) Verify and update documentation on add_compiler

documentation/add_compiler.md

@@ -0,0 +1,33 @@
+# Add Compiler
+
+The `peadm::add_compiler` plan can be used to add a new compiler to a PE cluster or replace an existing one with new configuration.
+
+## Add a compiler to an existing PE cluster
+
+Prepare to run the plan using a params.json file such as this one:
+
+```json
+{
+  "avail_group_letter": "A",
+  "compiler_host": "pe-xl-compiler-0.lab1.puppet.vm",
+  "dns_alt_names": [ "puppet", "puppet.lab1.puppet.vm" ],
+  "primary_host":  "pe-xl-core-0.lab1.puppet.vm",
+  "primary_postgresql_host": "pe-psql-6251cd-0.us-west1-a.c.slice-cody.internal",
+}
+```
+
+See the [install](install.md#reference-architectures) documentation for a list of supported architectures.
+
+
+## Running the add_compiler plan
+```
+bolt plan run peadm::add_compiler --params @params.json 
+```
+
+This command will retrieve the current PEADM configuration to determine the setup rules needed for a compiler's secondary PuppetDB instances. The plan will configure the primary with appropriate rules for allowing access from the new compiler. On the primary, the `puppet` service is stopped and the `pe-postgresql` service is reloaded. If required, a puppet agent is be installed. The compiler agent's certificate is be regenerated to include required data with `peadm::subplans::component_install`. Puppet agent will run on the following components
+* _\<compiler-host\>_
+* _\<primary_postgresql_host\>_
+* _\<replica host\>_
+* _\<primary_postgresql_host\>_
+
+ The `puppet` service is then restarted.

plans/add_compiler.pp

@@ -90,7 +90,7 @@
   # On <primary_postgresql_host> run the puppet agent
   run_task('peadm::puppet_runonce', $primary_postgresql_target)
 
-  # On replica puppetdb run the puppet agent
+  # On replica host run the puppet agent
   run_task('peadm::puppet_runonce', $replica_puppetdb_target)
 
   # On <primary_postgresql_host> start puppet.service