This feature will be built in to Chef 11. We've resisted adopting the easy solution of shoving data in the node because 1) we don't want to take a chance on putting bogus node data on the server and screwing up search results (could happen during a failed run); 2) This doesn't scale to keeping more than one run's data, so we knew we'd need to replace it eventually anyway; and 3) knowing we'd replace it, we didn't want to offer an API around which people would build tools only to completely change it later (we've changed APIs before, but it's not something we take lightly).
What we're implementing for Chef 11 will track client runs over time, including information about which resources are updated and how.
Hope that explains things a bit.
--
Dan DeLeo
Archive powered by MHonArc 2.6.16.