You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardexpand all lines: documentation/upgrade.md
+16-10
Original file line number
Diff line number
Diff line change
@@ -4,13 +4,15 @@ Puppet Enterprise deployments provisioned using the peadm module can also be upg
4
4
5
5
## Usage
6
6
7
-
The `peadm::upgrade` plan requires as input the version of PE to upgrade to, and the names of each PE infrastructure host. Primary, replica, compilers, etc.
7
+
The `peadm::upgrade` plan requires as input the version of PE to upgrade to, and the names of each PE infrastructure host. Primary, replica, compilers, etc.
8
8
9
-
The following is an example parameters file for upgrading an Extra Large architecture deployment of PE 2021.0.1 to PE 2021.7.4.
9
+
Please note that when upgrading from before 2023.3 to 2023.3 or above and you are using code manager, it is nessesary to provide known hosts for r10k. r10k_known_hosts is an optional parameter and is only required one time when upgrading to 2023.3 or beyond. Subsequent upgrades will already have this and wont be required again.
10
+
11
+
The following is an example parameters file for upgrading an Extra Large architecture deployment of PE 2023.2.0 to PE 2023.3.0.
@@ -115,20 +121,20 @@ Note: it is assumed that the Puppet primary is in cluster A when the upgrade sta
115
121
1. Shut down the `pe-puppetdb` service on the compilers in cluster B
116
122
2. If different from the primary (replica), run the `install-puppet-enterprise` script for the new PE version on the PuppetDB PostgreSQL node for cluster B
117
123
3. If different from the primary (replica), Run `puppet agent -t` on the PuppetDB PostgreSQL node for cluster B
118
-
5. Run `puppet agent -t` on the primary to ensure orchestration services are configured and restarted before the next steps
119
-
6. Perform the replica upgrade using `puppet infra upgrade replica` for the primary (replica)
120
-
7. Perform the compiler upgrade using `puppet infra upgrade compiler` for the compilers in cluster B
124
+
4. Run `puppet agent -t` on the primary to ensure orchestration services are configured and restarted before the next steps
125
+
5. Perform the replica upgrade using `puppet infra upgrade replica` for the primary (replica)
126
+
6. Perform the compiler upgrade using `puppet infra upgrade compiler` for the compilers in cluster B
121
127
122
128
**If Upgrading from 2019.5**
123
129
124
130
The following steps apply _only_ if upgrading from 2019.5 or older
125
131
126
132
1. Run `puppet infra run convert_legacy_compiler` for all compilers
127
133
2. Modify the peadm node groups "PE Compiler Group A" and "PE Compiler Group B" as follows:
128
-
* Re-parent the groups. They should be children of "PE Compiler"
129
-
* Remove configuration data (Hiera data). Leave the classes and class parameters
130
-
* Add the rule `trusted.extensions.pp_auth_role = pe_compiler`
131
-
* Remove the rule `trusted.extensions."1.3.6.1.4.1.34380.1.1.9812" = puppet/compiler`
134
+
* Re-parent the groups. They should be children of "PE Compiler"
135
+
* Remove configuration data (Hiera data). Leave the classes and class parameters
136
+
* Add the rule `trusted.extensions.pp_auth_role = pe_compiler`
137
+
* Remove the rule `trusted.extensions."1.3.6.1.4.1.34380.1.1.9812" = puppet/compiler`
# If the version is 2023.3 or greater and known_hosts is undef, print a message
10
+
if ($version_check and $r10k_known_hosts == undef) {
11
+
out::message("URGENT UPGRADE NOTICE:\nWhen you are upgrading to PE 2023.3 or later, it's crucial to ensure that the 'known_hosts' parameter for R10k is populated. Starting from PE 2023.3, Code Manager will not operate correctly without this. For more information, see https://puppet.com/docs/pe/2023.3/code_mgr_config.html#known-hosts-parameter")
0 commit comments