ah, i thought it was out already, or do you mean the docker image only?

I think Puppet doesnt maintain that image anymore

is there a reason that everyone but me is aware of 🙂 ?

they stopped maintaining all docker images IIRC

so not all

I believe they will continue to maintain the docker images again at some point of time later.. but not in Q3 🙂

This message was deleted.

@bastelfreak you mean publish the images somewhere in your space on dockerhub?

yes

This message was deleted.

This message was deleted.

This message was deleted.

This message was deleted.

This message was deleted.

puppet agents only pull down some difference - if you update your ssh login banner only, it should only pull that one thing down. Are you saying you cannot just change your login banner and see that one change, independently of,say, a new application config?

Hiera/ENC and profile parameters (“feature flags”) are your friends here

We've also made environments apps and basemodulepath infrastructure

so they can have parallel deployment pipelines

OK shameless plug for my book which comes out on 28th and everyone loves a free quiz.... Puppet and Packt are giving away 3 physical copies of Puppet 8 for DevOps engineers if you can answer some Puppet trivia :-) https://docs.google.com/forms/d/e/1FAIpQLSd8f8XKJ1AZS8uOrSrT-ACDHnm1dYwstaDSY2_65KQCrb7slA/viewform

beaker 🧑‍🏫Puppet Forge is about to start up in #CFD8Z9A4T

Hello, i have a question: how do you guys monitor that Puppet agent is running and doing it's job? From what i know there are two aspects: • service is not running (easy to handle in most monitoring software) • service is running, but agent is disabled via

puppet agent --disable "Reason XYZ"

For case when agent is disabled we've previously used "Monit" (service was running from root) which was checking last modification timestamp of the file

/opt/puppetlabs/puppet/cache/state/last_run_report.yaml

/etc/monit$ cat conf-enabled/puppetagent.monit 
### THIS FILE IS MANAGED BY PUPPET. ANY MANUAL CHANGES WILL GET OVERWRITTEN.
check file last_puppet_run with path /opt/puppetlabs/puppet/cache/state/last_run_report.yaml
  if mtime older than 1440 minutes then exec "/etc/monit/conf-enabled/slack_bad.sh"

We tried to try to replicate what Monit was doing in Zabbix and Prometheus node exporter, but because of the permissions of this file Zabbix and Prometheus node exporter are not able to check timestamp of this file

ls -l /opt/puppetlabs/puppet/cache/state/last_run_report.yaml
-rw-r----- 1 root root 284420 Jun 23 09:46 /opt/puppetlabs/puppet/cache/state/last_run_report.yaml

What is the proper way to cover use case when agent is disabled? Thanks

I am using icinga to monitor my infrastructure and basically i wrote a small script with sudo privileges that does exactly what you are doing. it reads the timestamp of the last run and compares it to the current system time. if the difference is too big (arbitrary value we set) it returns with an exit code of 1 or 2 instead of 0. The icinga agent calls this locally

If this is the best practice, i dont know

you can tell puppet to write the file into a different directory.

and if puppet is running as service you can monitor the systemd unit. I think the node_exporter supports that by default

you can also check for $(puppet config print agent_catalog_run_lockfile) and $(puppet config print agent_disabled_lockfile) to check if the agent is disabled / taking ages for a puppetrun

Thanks, will take a look at these options

You can monitor if agent is disabled by checking the lock file presence. Should be easy to use node_exporter text collector or telegraf for this.

or there is this option https://support.puppet.com/hc/en-us/articles/360006717334