On Tue, Jun 16, 2009 at 1:07 PM, Alexander Klimetschek<aklim...@day.com> wrote: > On Tue, Jun 16, 2009 at 7:55 AM, Felix Meschberger<fmesc...@gmail.com> wrote: >> ...Just to have a status mechanism for startup makes no sense. > > Well, the whole point of the system ready detection is to find out at > what point in time everything is up and running to serve requests...
> ...But I don't think that bundles going away or crashing during the > lifetime should lead to a 503 immediately. They might indicate a > problem and one wants to see this problem rather than a 503.... Agreed - and anyway I think this can be decided later, here's how I see it: The SystemStatus service will have to cache the "system is ready" info anyway, to avoid recomputing it for each request. If we want events (bundle installs etc) to cause the system to go back to the "not ready" mode, it's just a matter of clearing that cached state when such events happen. As long as the SystemStatus service provides a method to clear its cache, we can start with the "503 at startup" use case, and look at more sophisticated strategies later - if needed. -Bertrand
