Luke Kanies wrote:
> On Jul 24, 2009, at 3:48 AM, David Schmitt wrote:
> 
>> James Turnbull wrote:
>>> -----BEGIN PGP SIGNED MESSAGE-----
>>> Hash: SHA1
>>>
>>> David Schmitt wrote:
>>>> Even if that might mean having some .puppettmp file lying around  
>>>> from an
>>>> aborted filetransfer?
>>>>
>>> Isn't that implied by certain types of shut-downs?  Are you
>>> suggesting Puppet do some clean-up when someone Ctrl-C's for example?
>> "finishing the evaluation/application of the current resource" would
>> imply that. Yes. But that'll be probably type/provider specific. E.g.
>> Exec should really cancel the running command?
> 
> IMO, if you want a process to exit right now, you use SIGKILL.  In the  
> general case, Puppet should do its best to be safe.
> 
> It sounds a lot like HUP should finish transactions but TERM and INT  
> should only finish the current resource.  Any complaints about that  
> behaviour?
> 

HUP is normally reserved for reloading configurations, not terminating a 
process.  While an SA shouldn't blindly send signals to processes, I think 
having HUP actually terminate a process certainly violates the principle of 
Least Surprise.

Steven Jenkins
End Point Corporation

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Puppet Developers" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/puppet-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to