On Sunday, December 9, 2012 3:49:06 PM UTC-6, John Warburton wrote:
>
> On 8 December 2012 04:34, Peter Meier <peter...@immerda.ch 
> <javascript:>>wrote:
>
>>  And why shouldn't it?
>>
>
> Nagios is being 
> removed<https://groups.google.com/forum/?fromgroups=#!topic/puppet-users/qNxc2xZnRdM>
>  from 
> core <http://projects.puppetlabs.com/issues/4146>. Cfengine isn't core, 
> and falls under the same logic
>


Sort of.  The planned change to the nagios types is a refactoring, not a 
feature removal.  The nagios types will (maybe) be moved out of the core 
codebase, but they will still be packaged with Puppet.  Or maybe even that 
won't happen -- it was scheduled for Puppet 3.0, but it was deferred.

Furthermore, the Facter case is not parallel.  Few, if any, of the facts 
distributed with Facter are "core".  The Facter engine doesn't depend on 
them; instead they serve *Puppet's* purposes.

Let me not be completely negative.  It is fair to question which facts 
should be packaged with Facter, and I am inclined to think that Cfengine 
keys are included mostly for strategic reasons.  Every fact installed makes 
Facter take a little longer to run, so it's unreasonable to just throw in 
every fact that someone thinks would be useful.  On the other hand, a known 
set of facts need to be reliably present for Facter to serve its purpose.

In the end, however, Facter 2.x is going to keep the fact forever because 
of PL's commitment to semantic versioning.  Removing a feature would 
certainly constitute a breaking change, so even if PL were persuaded to do 
it, it would not happen until Facter 3 at the earliest.

What I would be thinking, therefore, is not whether Facter should include 
that fact, but whether its behavior is buggy.  If your system 
(mis)configuration caused Facter to fail while trying to evaluate the 
Cfengine key fact, then I would call that a bug in the fact (and file a 
ticket).  Of all software, Puppet and its supporting tools need to be among 
the most tolerant of broken and odd system configurations.


John

-- 
You received this message because you are subscribed to the Google Groups 
"Puppet Users" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/puppet-users/-/GVxhXUPVeyIJ.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to 
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/puppet-users?hl=en.

Reply via email to