https://www.puppet.com/community logo
Join Slack
Powered by
# voxpupuli
  • v
    I haven't had 140807021 online in a long time
  • v
    I just updated a bunch of servers from Puppet 6 to 7 and now I get deprecation warnings from Ruby 2.7; I'll say that running Puppet in cron mode and getting an email every x minutes is a good incentive to fix it …
  • v
    (gnupg module in case anyone is wondering - had already started on https://github.com/dgolja/golja-gnupg/pull/38)
  • v
    or use systemd timer 😎
    💯 1
  • v
    I am really hoping that puppet 8 support timeframe aligns with the ruby 3.2 EoL date
  • v
    doesn't really solve the point that I get an email for every Puppet run ;)
  • v
    well, if you migrate from cron to systemd, it will stop and it's an easy change for theforeman/puppet
  • d
  • v
    actually not a bad idea
  • v
    at least it means a quiet weekend and I can fix it properly next week
    ✔️ 1
  • v
    didnt Incontribute a timer to that module?
  • v
  • v
    it actually may be a good idea to submit the timer upstream so we don't need to carry a unit file in our repo anymore, but I never got around to that
  • v
  • v
    yes, I've neglected the puppet module PRs a bit
  • v
    vchepkov++ for the easy suggestion
  • d
    vchepkov gained a level! (Karma: 13)
  • v
    np. minimum efforts is my second name 🙂
  • p
    I'd say "neglect" is a strong word given you've looked at like 6 of my PRs against foreman/puppet this week alone
  • v
    if you were in marketing you'd call it time efficient solutions ;)
  • v
    well, the PR to fix linting was open for a long time
  • v
    so this week was mostly catching up on things
  • h
    I was doing new facts on facterdb and noticed that the 
    bundle exec rake table
    messes up some of the recent windows additions in the tables. I had to manually fix to keep from muddying the waters there
  • v
    would it make sense to exclude Facter < x from the table?
  • v
    I mean, how many people still care about Facter 1 and 2?
  • v
    it may even be fine to keep them in FacterDB (though that's also becoming questionable), but not display them
  • h
    I'd rather the sort order switch to be 4.2 for the first column and the 1.x as the right most column
  • v
    also something I thought about
  • l
    I would think that retiring factor 1 and 2 immediately makes sense given that 3 is finally going away with puppet 6 EOL
  • v
    there's also the runtime overhead of more fact sets (since it parses them all)
1...246247248...648Latest