Just some feedback from Charlie regarding the file function usage. "In general, reducing the number of file serving requests that hit JRubies is a good thing: https://tickets.puppetlabs.com/browse/PE-35439 When Code Manager is enabled, PE does this already for the

modules/

mount via static catalogs. Pulling content into the catalogs via

file()

is something you have. to. be. very. careful. with. Because that content is going to end up pushed out to places like PuppetDB and also increases the overhead of parsing and handling the JSON blob. It may work great for someone that has a handfull of

.pem

files and then someone with multi-megabyte files tries it and blows their entire leg off."

"another thing about

file()

is that it will always ship the content --- regardless of whether the agent actually needs to update the file. So, there's a persistent overhead there."

anyhoo I'm going to raise a doc ticket for us to try and get the subtly of the performance around file server and use of functions

anyhoo I'm going to raise a doc ticket for us to try and get the subtly of the performance around file server and use of functions

Puppetboard question - Looking at the puppetboard install module, it's not clear if I am to install/configure puppetDB FIRST, then install puppetboard, or just install it on top of a barebones puppetserver. It would appear it installs at least some of the puppetDB stuff, so maybe let it?

niiice..

Bundler.rubygems.find_name('puppet-blacksmith')

reports there is such module but there is no in Gemfile.lock

that’s part of PDK weirdness (single dir for everything)

not sure if there is better way.. definitely there is no problem when bundler is used directly 🙂 (non-PDK one I mean)

This message was deleted.

message has been deleted

how do you currently deploy the code on puppet6.domain.com?

I must admit, I have fallen into the bad habit of distributing some Go binaries using custom mount point from puppet server. Would it be advisable to instead switch to Nexus OSS (which I have available already) and

archive

? Largest file size is about 38MB.

This message was deleted.

Hey guys, where do I go for help with a puppet installation?

You may ask questions here or you may buy the commercial support, depending on the Puppet version (PE vs OSS)

I am using oss. I am having an issue with my codedir directory. I changed it to /srv/puppet instead of the default /etc/puppetlabs/code, however, the puppet agent appears to still be using the default directory.

I even used the command puppet config set codedir, however, I am still having the same issue. Does anyone know a way away round this?

This directory is used by puppetserver

If you’re using masterless puppet then you can apply the code from anywhere too though

Easiest way is to symlink the codedir maybe

I added 1 day ttl cache to most heavy facts (all from puppet/yum) and see a strange result

# facter -p -t yum_reboot_required|grep yum_reboot_required
cached fact 'yum_reboot_required', took: (0.000366) seconds
fact 'yum_reboot_required', took: (5.196689) seconds

Why is it cached and not cached at the same time ? bug ?

See the

puppet-yum

module at https://forge.puppet.com/puppet/yum?src=slack&channel=puppet

Is there something that one could add to metadata.json to require PuppetDB for a module? I have exported resources, I have a puppetdb query.

I think I can guess the answer, because I see nothing here:

<https://www.puppet.com/docs/puppet/7/modules_metadata.html>

I think I can guess the answer, because I see nothing here:

<https://www.puppet.com/docs/puppet/7/modules_metadata.html>

I wouldn’t expect there is a way tbh..

It’d be nice to have a way though.. maybe Content & Tooling team may consider this for future tasks.. cc @Craig Gumbley

This message was deleted.

Is there something like a puppet agent repo for windows server? A co-worker needs the agent for Windows 2022 Server (20348.1607) I mean there is apt.puppet.com and yum.puppet.com, but what is it for windows?

ah looks like downloads.puppetlabs.com/windows/