[Labs-l] Labs datacenter migration

Thu Dec 5 17:35:51 UTC 2013

On 11/20/13 12:31 PM, Diederik van Liere wrote:
> Would it be hard and worthwhile to write a 'migration assessment' 
> script that can be run on a labs instance and that would give you 
> warnings if you have data in your home folder, puppet does not run 
> properly, etc. Owners of a labs instance could then be notified by 
> email that their instance is probably not ready for migration. If this 
> would make sense, what kind of checks should such a script make?
I've been thinking about this off and on... so far I don't have a lot of 
good ideas about what the 'etc.' would consist of.  Checking an instance 
for 'important' data that's not in shared storage is non-trivial, so I'm 
left thinking that the main question here is one of whether or not 
puppet is running.

I've been tinkering with centralized puppet checks over the last few 
days... I may be able to generate a report that collates a list of 
instances with broken puppet.  I'd welcome your help with this -- just 
ping me on IRC or off list.