On Saturday, January 30, 2016 at 4:47:13 AM UTC+11, Rob Nelson wrote:
>
>
> While you can't force all the forge modules to be updated, I highly 
> suggest setting up rspec-puppet on your controlrepo. You'll find the 
> modules that don't support future parser with 3, or 4 in general, and file 
> tickets with the module authors. I had to do that for a number of modules I 
> use (ajjahn/dhcp and stahnma/epel ring a bell) and I found the authors are 
> generally receptive. If they're not, then it might be a good indicator that 
> you shouldn't be relying on that module; find another or fork it if the 
> license allows you to.
>
> Heck, maybe some module authors are watching this thread and are seeing 
> the demand for supporting Puppet 4 :)
>

This conversation has caused me to have another look at open source Puppet 
4; and I'll do that.

I think, however, that none of this changes the fact that Puppet 3 isn't 
going to magically disappear just because we all want it to.  Puppet 3.8.5 
was released just a week ago.  Puppet 2.7.26 was released October 2014.   
People will, in edge-cases, need to install a Puppet compatible Ruby via 
Yum, even in a Puppet 4 world.  Someone will find a reason to install 
Puppet 4 from a Gem on Ruby on EL6.

Why not create an incentive for the community to close the gaps in the 
Puppet Forge.  Draw up a list of gaps in the Puppet Forge, and offer $1000 
to the first community developer who can produce an Approved Puppet 3 & 4 
compatible module for each Gap.  It'd be petty cash to Puppet Labs, and 
I'll bet we'd close out all the gaps in the Forge in 3 months.  I'll 
certainly play. ;-)

-- 
You received this message because you are subscribed to the Google Groups 
"Puppet Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/puppet-dev/bb64c970-e33a-477e-8b33-4b22fb42f678%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to