You still don't, code manager does

it’s remains true, you don’t change anything there

all the fun stuff is there! 😄

yeah, but my stuff doesn’t spectacularly crash like some people I know. ಠ_ಠ

The

puppet-agent

package creates

/etc/puppetlabs/code/environments/production

along with files containing defaults to help users get started. However, when lockless code deploy is enabled, puppetserver's

codedir

is configured to point to the new directory.

if you're going to move a core system envirnment location, perhaps a grep through all the codebases

Some people run serverless, so

puppet-agent

has to create the old directory

Seems like something to let Puppet know what's up such as maybe a new puppet.conf setting: install_type = serverless and then leverage that as a core setting that sort of directs traffic would be an idea I'd offer. That way you're not pooping extra directories onto disk if they've been deprecated or moved? Just thinking out loud.

except non-PE uses the other dir, and isn't changing... and the agent isn't different between the two?

my colleague working on a PE upgrade ran into this https://puppet.atlassian.net/browse/DOCUMENT-1230

Hello all, We have our puppet master set up behind a DNAT on NSX edge. What would be the right way to set up compilers to use DNAT for all neccesery services AND is such approach is actually supported? In current documentation it suggested that compile masters talk directly to the Master server. Using

/etc/hosts

on compilers is not the solution for our case since we usually forget about that one when migrating stuff, and it is more of a hack than a solution. Is it possible at all to set up and configure all the necessary services to use DNAT address in puppet console? Thank you!

Names have to be in place and working no matter what. However you decide to address, you will still need either DNS entries or Host File entries to point to the address you use without exception.

Using the hosts file for names is not really a hack in cloud situations, as multiple interfaces and routes come into play, and your overall design needs to fit within the well architeted framework considerations for your particular cloud provider. If on-prem, simply expose an IP, and point the name(s) in question to the cpnfigured IP, and things should function. Don't sell host file entries short, as they insert into the lib gethostbyname() architecture, and are a perfectly viable alternative if you do not have access to DNS configurtion

I need a new keyboard. LOL

Can bolt be used to do puppet agent upgrades like we use it for pe upgrade

Use the Puppet_agent module and just do it through desired state. The agent is just a package, so you could use a 'bolt command run' or 'bolt task run package'.

I wonder, is augeas still the best way to make sure a config file has a certain config line in it? In short, are there any better (easier) ways to achieve the same without the complex syntax of augeas?

File: /etc/puppetlabs/code/environments/production/modules/puppet_agent/manifests/install/windows.pp

Getting failure on puppet run - agent upgrade failure

Any suggestions on how to resolve this?

Why error do you get and which command did you run?

Error: Unable to connect to server from server_list setting: Request to https://us03p02pup10.prod.nexttravel.com:8140/status/v1/simple/server failed after 0.001 seconds: Failed to open TCP connection to us03p02pup10.prod.costcotravel.com:8140 (Connection refused - connect(2) for "us03p02pup10.prod.nexttravel.com" port 8140)

What is the best way to resolve this error?

For some reason puppet broke

Hello all, Do we know when will PE 2023.8.0 will be available for Ubuntu 24 ? What's the plan ? Thanks in advance,

On the pe primary puppet agent should run ? systemctl status puppet * puppet.service - Puppet agent Loaded: loaded (/usr/lib/systemd/system/puppet.service; enabled; vendor preset: disabled) Active: inactive (dead) since Thu 2024-08-01 042329 UTC; 3 months 4 days ago Docs: man:puppet-agent(8)

Module sync is not happening from jfrog artifactory

sudo -H -u pe-puppet bash -c 'curl -X GET -H "Authorization: Bearer {Token}" https://artifactory url/artifactory/api/puppet/puppet-prod-virtual/v3/modules/filebeat' List all the modules correctly. Mentioned the specific version module on Puppetfile control repo for specific environment. it is not updating. Any idea? Still taking oold module version.

the puppet unit is for the puppet agent itself, all the PE specific services are prefixed with pe-

ok. All the pe- services are works well. systemctl status pe-puppetdb systemctl status pe-console-services systemctl status pe-puppetserver

okay