https://www.puppet.com/community logo
Join Slack
Powered by
# puppet
  • n
    if its while I am at work, yes I use vmware and snapshots. If its outside of work which I do some for also, then I have terraform and Azure
  • u
    You just delete the VM and re-create it using Terraform?
  • n
    on Azure yes
  • n
    I have it all scripted. it builds and joins my puppet infra in minutes
    đź‘Ť 1
  • s
    This message was deleted.
    y
    n
    +2
    • 5
    • 11
  • y
    AWS's git product (codecommit) sucks so bad.
  • a
    hi all
  • a
    Copy code
    yumrepo { "epel$operatingsystemmajrelease-x86_64":
    ensure   => present,
    descr    => "epel$operatingsystemmajrelease-x86_64",
    baseurl  => "<http://mydomain.com/epel/$operatingsystemmajrelease/x86_64|mydomain.com/epel/$operatingsystemmajrelease/x86_64>",
    enabled  => 1,
    gpgcheck => 0,
    priority => 99,
    proxy    => $::common::repos::yumproxy,
  }
  • s
    This message was deleted.
    t
    v
    +2
    • 5
    • 38
  • s
    This message was deleted.
    âž• 1
    y
    • 2
    • 1
  • v
    maybe some local fact overshadows system one?
  • r
    <rant> why does no one use semantic versioning </rant
    ```Gem:RuntimeRequirementNotMetError rubocop-ast requires Ruby version >= 2.6.0.
    The current ruby version is 2.5.9.229.
    An error occurred while installing rubocop-ast (1.18.0), and Bundler cannot
    continue.```
  • r
    I mean, I know Ruby 2.5 is technically end of life, but still
  • b
    I think this is more of an issue for an olddates rubygems verison that cannot handle the mentioned Ruby version in .gemspec files
  • r
    So I could theoretically update rubygems in my pipeline to fix??
  • c
    ruby is vendored, so I would imagine that any changes at all to /opt/puppetlabs would make your support….. questionable.
  • c
    especially the ruby home
  • r
    in this case it’s just a CI pipeline that is running generic Ruby 2.5.x
  • r
    match the version of Ruby with the version of ruby that is vendor in the agent. So all of our Puppet 6.x pipelines are failing due to Ruby 2.5
  • v
    Hopefully LTS is out soonish and everyone starts migrating to puppet 7
  • r
    Yeah. It’s slightly frustrating to deal with, but usually the Vox Pupuli catch the issue in their pipelines before I do, so admittedly the work on my end has been very little
  • r
    Honestly best to ignore this without logs. Can't help without relevant information
  • b
    yep!
  • b
    puppet 7 works well on ruby 3, if you like to use that
    🎉 1
  • b
    and yes, updating your rubygems should fix this error
  • v
    hmm, but it uses 2.7, right?
  • b
    yes thats vendored and oooold :(
  • r
    So could be due to some weird auto require for example
  • s
    This message was deleted.
    b
    o
    • 3
    • 9
  • o
    Hello everyone, I would like to ask your opinion on migration to Puppet7. Right now we have Puppet 5 with +-300 nodes. How would you do it? What is your vision of propper and technically correct migration? I will be doing it for the first time. So far my plan is next: 1. Install new OS + new Puppet 2. Fork old "control-repo" to new "control-repo7" 3. In old control-repo --> create new branch (with only 1 change: replace old address of Puppet server to new host) 4. Apply branch (pointed to new Puppet server) -->> and them migrate one by one So far one of the questions that we have -> how to do it in "easy" but in the same time correct way? For example I'm thinking if it's possible to export old Puppet CA key + certificate, and re-use them (import CA) in new Puppet 7 host, and then also transfer SSL certs for currently used nodes from Puppet 5 to Puppet 7 host so we might in theory switch back and forth without need to remove old certificates and then re-create them in new Puppet. Is this good idea overall? Or better to not use old CA?
1...404142...428Latest