Cool. Thanks for the explanation!
On Thu, Dec 5, 2013 at 6:08 PM, Daniel DeLeo email@example.com wrote:
On Thursday, December 5, 2013 at 2:47 PM, Tim Dunphy wrote:
I’ve recently setup a little test lab with chef intended to control
I’m running chef 11.8.2 on an ubuntu 13.10 server.If I run chef-client
manually on the hosts that are falling behind, it catches up with the
server and the web UI in on the chef server reports that they are in sync.
But if I leave it alone for a while this happens again, until I run it
manually. Short of setting up a cron job to keep the hosts in sync, is
there a more elegant solution to this problem? How do I troubleshoot this
from a chef perspective?
The status page assumes you’re running chef either daemonized or from
cron, and something is wrong if chef hasn’t run in a while. If you agree
with that, then you should run chef daemonized or via cron. If you have a
good reason not to do that, then you should agree to disagree with what the
status page is telling you and ignore it.
FWIW, the status page is kind of a primitive attempt at a reporting system
from long before Opscode had enough engineering resources to write such a
thing. The paid Chef server versions are getting a much more advanced
reporting system now:
If you’d prefer to stick with the OSS version, you can integrate Chef with
a different system via report handlers:
Or using the event stream (which is what the client-side code for the paid
reporting feature does).
gpg --keyserver pool.sks-keyservers.net --recv-keys F186197B